CN108132717A - Recommendation method, apparatus, storage medium and the mobile terminal of candidate word - Google Patents
Recommendation method, apparatus, storage medium and the mobile terminal of candidate word Download PDFInfo
- Publication number
- CN108132717A CN108132717A CN201711391875.9A CN201711391875A CN108132717A CN 108132717 A CN108132717 A CN 108132717A CN 201711391875 A CN201711391875 A CN 201711391875A CN 108132717 A CN108132717 A CN 108132717A
- Authority
- CN
- China
- Prior art keywords
- candidate word
- recommended
- word
- candidate
- input
- 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.)
- Pending
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
- G06F3/0237—Character input methods using prediction or retrieval techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3322—Query formulation using system suggestions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Abstract
The embodiment of the present application discloses recommendation method, apparatus, storage medium and the mobile terminal of a kind of candidate word.This method includes:It obtains current session above, and candidate word recommended models is input to above by described, wherein, the candidate word recommended models are to be trained based on word dialog record as training sample using machine learning means;According to the candidate word recommended models, candidate word to be recommended is determined;The candidate word to be recommended is shown in the candidate word display area of input interface.Technical solution provided herein can realize the effect for the recommendation applicability for improving candidate word.
Description
Technical field
The invention relates to the recommendation method, apparatus of field of communication technology more particularly to a kind of candidate word, storages to be situated between
Matter and mobile terminal.
Background technology
With the development of mobile communication technology, people are linked up commonly using electronic equipment, such as soft by some chats
Part is chatted.
At present, in input method, user is determined as the continuous input sequence of word often by counting user and is pushed away
The reference weight for the candidate word recommended, such as user by continuously inputting " ten give power " after, when reuse the input method input " ten
Point " when, then it in candidate word " gives power " and is likely to make number one, however this method and not smart enough, many times user
User can not be chosen in candidate word needs word to be used, causes the waste of candidate word recommendation function, while also cause
The experience of user is bad, needs to improve.
Invention content
The embodiment of the present application provides a kind of recommendation method, apparatus, storage medium and the mobile terminal of candidate word, can realize
Improve the effect of the recommendation applicability of candidate word.
In a first aspect, the embodiment of the present application provides a kind of recommendation method of candidate word, this method includes:
It obtains current session above, and candidate word recommended models is input to above by described, wherein, the candidate word pushes away
It is to be trained based on word dialog record as training sample using machine learning means to recommend model;
According to the candidate word recommended models, candidate word to be recommended is determined;
The candidate word to be recommended is shown in the candidate word display area of input interface.
Second aspect, the embodiment of the present application provide a kind of recommendation apparatus of candidate word, which includes:
MIM message input module for obtaining current session above, and is input to candidate word recommended models above by described,
Wherein, the candidate word recommended models are as training sample, using the training of machine learning means based on word dialog record
Into;
Candidate word determining module to be recommended, for according to the candidate word recommended models, determining candidate word to be recommended;
Candidate word recommending module, for showing the candidate word to be recommended in the candidate word display area of input interface.
The third aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence realizes the recommendation method of the candidate word as described in the embodiment of the present application when the program is executed by processor.
Fourth aspect, the embodiment of the present application provide a kind of mobile terminal, including memory, processor and are stored in storage
And can be in the computer program of processor operation on device, the processor is realized when performing the computer program as the application is real
Apply the recommendation method of the candidate word described in example.
The technical solution that the embodiment of the present application is provided inputs above by obtaining current session above, and by described
To candidate word recommended models, wherein, the candidate word recommended models are to utilize machine as training sample based on word dialog record
Device learning ways are trained;According to the candidate word recommended models, candidate word to be recommended is determined;In the candidate word of input interface
Display area shows the candidate word to be recommended, can realize the effect for the recommendation applicability for improving candidate word.
Description of the drawings
Fig. 1 a are a kind of flow diagram of the recommendation method of candidate word provided by the embodiments of the present application;
Fig. 1 b are a kind of recommendation schematic diagram of candidate word provided by the embodiments of the present application;
Fig. 2 a are the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application;
Fig. 2 b are the recommendation schematic diagram of another candidate word provided by the embodiments of the present application;
Fig. 3 a are the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application;
Fig. 3 b are the recommendation schematic diagram of another candidate word provided by the embodiments of the present application;
Fig. 4 is the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application;
Fig. 5 is the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application;
Fig. 6 is a kind of structure diagram of the recommendation apparatus of candidate word provided by the embodiments of the present application;
Fig. 7 is a kind of structure diagram of mobile terminal provided by the embodiments of the present application.
Specific embodiment
Further illustrate the technical solution of the application below with reference to the accompanying drawings and specific embodiments.It is appreciated that
It is that specific embodiment described herein is used only for explaining the application rather than the restriction to the application.It further needs exist for illustrating
, part relevant with the application rather than entire infrastructure are illustrated only for ease of description, in attached drawing.
It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail
The processing described as flow chart or method.Although each step is described as the processing of sequence, many of which by flow chart
Step can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of each step can be rearranged.When its operation
The processing can be terminated during completion, it is also possible to have the additional step being not included in attached drawing.The processing can be with
Corresponding to method, function, regulation, subroutine, subprogram etc..
Fig. 1 a are a kind of flow diagram of the recommendation method of candidate word provided by the embodiments of the present application, and this method can be by
The recommendation apparatus of candidate word performs, and wherein the device can generally be integrated in the terminal by software and or hardware realization.
As shown in Figure 1a, this method includes:
S101, it obtains current session above, and candidate word recommended models is input to above by described, wherein, the time
It is to be trained based on word dialog record as training sample using machine learning means to select word recommended models.
Wherein, current session can be above conversation content in the communication software of mobile terminal above, can also
It is in mhkc, circle of friends etc. leave a message interface, the mutual message between other side and user that user's needs are replied can be got
As above.What deserves to be explained is signified in the embodiment of the present application can be voice messaging above or word is believed
Breath, if voice messaging, can be by speech-to-text after, obtain the word content of the voice messaging.
In the embodiment of the present application, the candidate word recommendation being previously mentioned can detect that user opens certain application program and at this
When input method progress word input is called in application program, candidate word is provided to the user, so that recommendation of the user according to candidate word
As a result, carrying out the selection to candidate word, the input time of the information of user is saved, and also is able to reduce user to input method
Algorithm burden caused by pressing indirectly saves the memory headroom of terminal.
Wherein, the candidate word recommended models can utilize engineering as training sample based on word dialog record
Habit means are trained.Wherein, word dialog record can be a large amount of conversation content got from network, engage in the dialogue
A large amount of network dialogue content is trained, obtains candidate word recommended models by Content Selection.Wherein, word dialog record is gone back
It can be that word dialog caused by communication software used in client user records, and then by training, obtain and user
The relevant candidate word recommended models of language communicative habits, the candidate word recommendation results that can cause are more accurate, and accord with
Share the word use habit in the chat process of family.The embodiment of the present application can also be realized by using candidate word recommended models
It provides to the user and is better able to the candidate word for showing user view, user can be assisted to complete the hint expression of oneself, avoided
User can not be gone out the thought of oneself by the literal expression inputted completely when ability to express is not strong, can improve user
Written communication during experience effect.
S102, according to the candidate word recommended models, determine candidate word to be recommended.
By current session be input to candidate word recommended models above after, can be obtained waiting to push away by candidate word recommended models
Recommend candidate word.Wherein, candidate word to be recommended can be one or multiple, can be with when candidate word is multiple
Recommend according to the recommender score of candidate word to be recommended, wherein, mould is recommended according to current session above and according to candidate word
Type, the recommender score that can more meet the candidate word of the intention of user are higher.
In the prior art, candidate word to be recommended is often determined according to current input content, and the application is real
Applying technical solution that example provided can be when user input any content, directly according to the progress above of current session candidate
Word is recommended, and the benefit set in this way is can quickly to provide the content that user wants input to the user, expands what candidate word was recommended
The scope of application, and improve the conversion ratio that candidate word is converted into input content.
S103, the candidate word display area displaying candidate word to be recommended in input interface.
Fig. 1 b are a kind of recommendation schematic diagram of candidate word provided by the embodiments of the present application.As shown in Figure 1 b, when receive work as
Preceding dialogue above when, i.e. " being fond of eating to the apple that you take " that " auntie " in Fig. 1 b says user can be interior above by this
Appearance is input to candidate word recommended models, can both obtain at least one candidate word to be recommended, and in the candidate word of input interface
Display area shows candidate word to be recommended, such as " the special sweet tea of apple ", " apple ", " especially nice " and " very sweet tea " in Fig. 1 b
Deng.
In the schematic diagram given by the present embodiment, it can be seen that when user is not in input interface any content of input,
It remains able to show candidate word to be recommended in candidate word display area, and the content recommended not is defeated according to user
The content entered, but according to the determining above of current session.
What deserves to be explained is current session signified in the embodiment of the present application is not limited in current session other side above
Multiple sentences that the previous sentence or conversation partner sent out is sent out, can according in candidate word recommended models for
The reply result of multiple sentences carries out candidate word recommendation or carries out multiple statement semantics of current session other side whole
It closes, so as to obtain candidate word recommendation results, and shows user.And in the embodiment of the present application, current session above may be used
To be not limited only to interrogative sentence or declarative sentence, the candidate word recommended is also not limited to the answer to the interrogative sentence, can be with
It is further extending or be switched to above other related or incoherent topics to the topic.
The technical solution that the embodiment of the present application is provided inputs above by obtaining current session above, and by described
To candidate word recommended models, wherein, the candidate word recommended models are to utilize machine as training sample based on word dialog record
Device learning ways are trained;According to the candidate word recommended models, candidate word to be recommended is determined;In the candidate word of input interface
Display area shows the candidate word to be recommended, can realize the effect for the recommendation applicability for improving candidate word.
Fig. 2 a are the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application, such as Fig. 2 a institutes
Show, this method includes:
S201, it obtains current session above, and candidate word recommended models is input to above by described, wherein, the time
It is to be trained based on word dialog record as training sample using machine learning means to select word recommended models.
S202, according to the candidate word recommended models, determine primary candidate word to be recommended.
In the embodiment of the present application, it is existing according to candidate word recommended models to provide one kind, obtains primary candidate word to be recommended
Afterwards, in the scheme screened again.
S203, the button operation for obtaining user, according to the correspondence of the button operation and the candidate word to be recommended,
The candidate word to be recommended is screened again.
Wherein, the button operation of user is obtained, can detect void of the user for position corresponding to the contact of touch screen
Intend button or actual key, wherein, button operation can be the input operation for text information, as nine grids are pressed
Such as the specific button of 1-9 in key, other buttons, such as comma button, fullstop button and Chinese and English switching key can also be
Deng, wherein, when user press be symbol keys or input mode switching key when, can be accordingly by primary time to be recommended
Word is selected further to be screened.In the present embodiment, specifically, can switch to English button when user presses Chinese
When, primary candidate word to be recommended is screened, the English word in primary candidate word to be recommended is recommended.It sets in this way
Benefit be that can provide the candidate word of actual needs to the user according to the use demand of user, so as to improve user use body
It tests.
Fig. 2 b are the recommendation schematic diagram of another candidate word provided by the embodiments of the present application.As shown in Figure 2 b, when receiving
Current session above when, then can be in original primary candidate to be recommended and when user presses " 8tuv " this button
It is further screened in word " the special sweet tea of apple ", " apple ", " especially nice " and " very sweet tea ".It will be write with carrying phonetic head
The candidate word that " especially nice " of female " t " beginning is recommended as override, will contain " very sweet tea " conduct of initial letter " t " beginning
The candidate word of second preferential recommendation is moved after the recommendation order of other candidate words.What deserves to be explained is in the embodiment of the present application not
Only with four candidate words as an example, can be understood as other times to be recommended in the technical program and a upper technical solution
Select recommendation order of the word in two technical solutions after this four candidate words.The benefit set in this way is can be according to obtaining
The information input operation of the user got is treated recommended candidate word and is adjusted correspondingly, and then improves turn that candidate word is selected
Rate, and the usage experience of user can be improved.
S204, the candidate word display area displaying candidate word to be recommended in input interface.
The technical program based on the above technical solution, provides a kind of button operation according to user to be recommended
The method that candidate word is further screened, not only can be according to the carry out candidate word above of current session from the present embodiment
Recommend, additionally it is possible to which the content for actually wanting to input to user according to the button operation of user is predicted, improves the time of recommendation
The applicability of word is selected, and the usage experience of user can be improved.
Fig. 3 a are the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application, such as Fig. 3 a institutes
Show, this method includes:
S301, it obtains current session above, and candidate word recommended models is input to above by described, wherein, the time
It is to be trained based on word dialog record as training sample using machine learning means to select word recommended models.
S302, according to the candidate word recommended models, determine primary candidate word to be recommended.
In the embodiment of the present application, it is existing according to candidate word recommended models to provide one kind, obtains primary candidate word to be recommended
Afterwards, in the scheme screened again.
S303, input content of the user in dialog box is obtained, according to the input content and the time to be recommended
The matching relationship of word is selected, the candidate word to be recommended is screened again.
Wherein, input content can be the content for being input to dialog box, be pushed away according to the content inputted with waiting
The matching relationship of candidate word is recommended, wherein matching relationship can be semantic matches, part of speech matching etc. be can also be, for example, having inputted
Content then can determine the word with turnover semanteme in candidate word to be recommended if " there are no eat " according to semantic matches,
Can such as use " but ", to bring turnover sentence into.According to part of speech matching can be when the content inputted for " very " when, it is existing
Candidate word no matter what the content of front is, can all recommend " good ", " thanks ", " liking " and " liking " etc., and in this Shen
Please in, can will be carried out in the candidate word determining above according to dialogue with the candidate word that " very " this part of speech is mutually arranged in pairs or groups
Recommend, such as adjective+noun, adverbial word+verb etc., and then obtain the candidate words such as " sweet tea " or " nice ".
Fig. 3 b are the recommendation schematic diagram of another candidate word provided by the embodiments of the present application.As shown in Figure 3b, when receiving
Current session above when, can according to the content inputted, and then determine candidate word to be recommended, such as currently inputted
When " especially nice, ", then it can determine that user will individually lift one section of word according to punctuation mark, according to having inputted
Appearance is " especially nice ", has been completed the reply above to conversation partner substantially, then this topic can be continued extension or
Person is converted to other topics, therefore, can obtain candidate word to be recommended as " apple of what kind ", " mouthfeel " and " very
It is clear and melodious " etc. extensions or topic shift candidate word.The benefit set in this way is can be according to the content inputted to candidate
After word screening, candidate word and the content similar import that has inputted or identical are avoided, influences the usage experience of user, Neng Gouzhen
The meaning of the positive content inputted according to user determines that user wants the content of input, reaches the mesh of intelligent recommendation candidate word
's.
S304, the candidate word display area displaying candidate word to be recommended in input interface.
The technical program provides a kind of input content according to user and treats on the basis of above-mentioned each technical solution
The method that recommended candidate word is further screened, not only being waited above according to current session from the present embodiment
Word is selected to recommend, additionally it is possible to which the content for actually wanting to input to user according to the input content of user is predicted, is improved and is pushed away
The applicability for the candidate word recommended, and the usage experience of user can be improved.
On the basis of above-mentioned each technical solution, optionally, further include:Obtain the word dialog record conduct that history generates
Training sample using machine learning means, determines in the word dialog record foregoing description and the hereinafter incidence relation of word;
According to the incidence relation, word model hereinafter corresponding with foregoing description is determined, as candidate word recommended models.
Wherein, according to the incidence relation of foregoing description and word hereinafter, candidate word recommended models are determined, it can be more accurate
Get users for the reply mode of foregoing description or get reply party of the active user for foregoing description
Formula, and then reply mode of the user for foregoing description is obtained, and provided to the user according to the model for largely learning to obtain and wait to push away
Recommend candidate word.The benefit set in this way is can to improve the applicability of recommended candidate word, can be according to learning outcome to waiting
Word is selected to be recommended, improves the conversion ratio that user uses candidate word.
On the basis of above-mentioned each technical solution, optionally, current session is obtained above, and be input to described above
The candidate word recommended models, including:Obtain the time of origin above of the current session;When time of origin is in default
Described it is input to the candidate word recommended models above in section.Wherein, preset period of time can be 10 minutes, can also be 5 points
Clock or shorter and longer time, the benefit set in this way are can be directed to avoid the candidate word recommended for user
It had occurred and that prolonged conversation content, and wanted to select word so as to show that some users are not conceivable when content inputs
Content can improve the laminating degree of recommended candidate word and user view, and then user is for the acceptance level of candidate word.
Fig. 4 is the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application, as shown in figure 4,
This method includes:
S401, the application environment for obtaining the input of user's word.
Wherein, the application environment can be some application program or certain a kind of application program, apply
The type of program can classify according to the usage type of application program, such as communication class, map class.
S402, corresponding nondeclarative memory dictionary is established according to the application environment.
It is corresponding to establish nondeclarative memory dictionary corresponding with application environment.
S403, when detect user open destination application when, determine program corresponding with the destination application
Phrase library for temporary memory.
S404, it obtains current session above, and candidate word recommended models is input to above by described.
S405, in described program phrase library for temporary memory, according to the candidate word recommended models, determine candidate word to be recommended.
S406, the candidate word display area displaying candidate word to be recommended in input interface.
The technical program additionally provides the difference according to application program, Huo Zheying on the basis of above-mentioned each technical solution
With the difference of the type of program, the candidate word nondeclarative memory dictionary corresponding to application program or Application Type is established,
Candidate word recommendation is carried out from candidate word nondeclarative memory dictionary can greatly improve candidate word constant speed degree really, be waited so as to improve
The advisory speed of word and display speed are selected, improves the usage experience of user.
Fig. 5 is the flow diagram of the recommendation method of another candidate word provided by the embodiments of the present application, as shown in figure 5,
This method includes:
S501, dialogue is obtained above, according to the exchange way for determining user and conversation partner above.
Wherein, can be the length according to two-party conversation content according to the exchange way for determining user and conversation partner above
It is short and as etc. the corresponding exchange way of foundation and both sides relationship, such as when the relationship between user of conversation partner is long
Whens generation, friend, higher level and classmate etc., different relationship types can be determined according to different exchange ways, and establishes and is somebody's turn to do
Certain memory dictionary corresponding to relationship type.
S502, corresponding certain memory dictionary is established according to the exchange way.
It is corresponding to establish certain memory dictionary corresponding with the exchange way.
S503, when detecting the exchange way of user and current session other side there are during corresponding certain memory dictionary,
In the certain memory dictionary, according to the candidate word recommended models, candidate word to be recommended is determined.
Wherein, dialogue is inputted into the candidate word recommended models above, it can be according to candidate word recommended models in the spy
Determine to determine candidate word to be recommended in phrase library for temporary memory.
S504, the candidate word display area displaying candidate word to be recommended in input interface.
The technical program provides on the basis of above-mentioned each technical solution and a kind of determines certain memory according to exchange way
Dictionary determines candidate word to be recommended further according to candidate word recommended models, and the benefit set in this way is can to improve recommended time
The matching degree of word and current exchange way is selected, improves the probability that recommended candidate word is used.
Fig. 6 is a kind of structure diagram of the recommendation apparatus of candidate word provided by the embodiments of the present application, which can be by software
And/or hardware realization, it is typically integrated in mobile terminal, candidate word can be pushed away by performing the recommendation method of candidate word
It recommends.As shown in fig. 6, the device includes:
The candidate word that is input to above for obtaining current session above, and is recommended mould by MIM message input module 601
Type, wherein, the candidate word recommended models are to be trained based on word dialog record as training sample using machine learning means
It forms;
Candidate word determining module 602 to be recommended, for according to the candidate word recommended models, determining candidate word to be recommended;
Candidate word recommending module 603, for showing the candidate word to be recommended in the candidate word display area of input interface.
The technical solution that the embodiment of the present application is provided inputs above by obtaining current session above, and by described
To candidate word recommended models, wherein, the candidate word recommended models are to utilize machine as training sample based on word dialog record
Device learning ways are trained;According to the candidate word recommended models, candidate word to be recommended is determined;In the candidate word of input interface
Display area shows the candidate word to be recommended, can realize the effect for the recommendation applicability for improving candidate word.
On the basis of above-mentioned each technical solution, optionally, the candidate word determining module 602 to be recommended, including:
Primary candidate word determination unit to be recommended, for according to the candidate word recommended models, determining primary time to be recommended
Select word;
First screening unit again, for obtaining the button operation of user, according to the button operation with it is described to be recommended
The correspondence of candidate word screens the candidate word to be recommended again.
On the basis of above-mentioned each technical solution, optionally, according to the candidate word recommended models, candidate to be recommended is determined
Word, including:
Primary candidate word determination unit to be recommended, for according to the candidate word recommended models, determining primary time to be recommended
Select word;
First screening unit again, for obtaining input content of the user in dialog box, according in described inputted
Hold the matching relationship with the candidate word to be recommended, the candidate word to be recommended is screened again.
On the basis of above-mentioned each technical solution, optionally, described device further includes:
Incidence relation determining module for obtaining the word dialog record of history generation as training sample, utilizes machine
Learning ways determine in the word dialog record foregoing description and the hereinafter incidence relation of word;
Candidate word recommended models determining module, for according to the incidence relation, determine it is corresponding with foregoing description hereinafter
Word model, as candidate word recommended models.
On the basis of above-mentioned each technical solution, optionally, described information input module 601, including:
Time of origin acquiring unit above, for obtaining the time of origin above of the current session;
Input unit above pushes away for being input to the candidate word above described in time of origin is in preset period of time
Recommend model.
On the basis of above-mentioned each technical solution, optionally, further include nondeclarative memory dictionary and establish module, be specifically used for:
Obtain the application environment of user's word input;
Corresponding nondeclarative memory dictionary is established according to the application environment;
When detecting that user opens destination application, nondeclarative memory word corresponding with the destination application is determined
Library;
In described program phrase library for temporary memory, according to the candidate word recommended models, candidate word to be recommended is determined.
On the basis of above-mentioned each technical solution, optionally, further include certain memory dictionary and establish module, be specifically used for:
Dialogue is obtained above, according to the exchange way for determining user and conversation partner above;
Corresponding certain memory dictionary is established according to the exchange way;
When detecting the exchange way of user and current session other side there are during corresponding certain memory dictionary, in the spy
Determine in phrase library for temporary memory, according to the candidate word recommended models, determine candidate word to be recommended.
The said goods can perform the method that any embodiment of the present invention is provided, and have the corresponding function module of execution method
And advantageous effect.
The embodiment of the present application also provides a kind of storage medium for including computer executable instructions, and the computer can perform
When being performed by computer processor for performing a kind of recommendation method of candidate word, this method includes for instruction:
It obtains current session above, and candidate word recommended models is input to above by described, wherein, the candidate word pushes away
It is to be trained based on word dialog record as training sample using machine learning means to recommend model;
According to the candidate word recommended models, candidate word to be recommended is determined;
The candidate word to be recommended is shown in the candidate word display area of input interface.
Storage medium --- any various types of memory devices or storage device.Term " storage medium " is intended to wrap
It includes:Install medium, such as CD-ROM, floppy disk or magnetic tape equipment;Computer system memory or random access memory, such as
DRAM, DDR RAM, SRAM, EDO RAM, blue Bath (Rambus) RAM etc.;Nonvolatile memory, such as flash memory, magnetic medium
(such as hard disk or optical storage);Memory component of register or other similar types etc..Storage medium can further include other
The memory or combination of type.In addition, storage medium can be located at program in the computer system being wherein performed or
It can be located in different second computer systems, second computer system is connected to computer by network (such as internet)
System.Second computer system can provide program instruction and be used to perform to computer.Term " storage medium " can include can
To reside in different location two or more storage mediums of (such as in different computer systems by network connection).
Storage medium can store the program instruction (such as being implemented as computer program) that can be performed by one or more processors.
Certainly, a kind of storage medium for including computer executable instructions that the embodiment of the present application is provided, computer
The recommendation operation for the candidate word that executable instruction is not limited to the described above, can also be performed what the application any embodiment was provided
Relevant operation in the recommendation method of candidate word.
The embodiment of the present application provides a kind of mobile terminal, and time provided by the embodiments of the present application can be integrated in the mobile terminal
Select the recommendation apparatus of word.Fig. 7 is a kind of structure diagram of mobile terminal provided by the embodiments of the present application.As shown in fig. 7, the shifting
Dynamic terminal can include:Memory 701, central processing unit (Central Processing Unit, CPU) 702 (are also known as handled
Device, hereinafter referred to as CPU), circuit board (not shown) and power circuit (not shown).The circuit board is placed in described
The space interior that housing surrounds;The CPU702 and the memory 701 are arranged on the circuit board;The power circuit,
For powering for each circuit or device of the mobile terminal;The memory 701, for storing executable program code;
The CPU702 is run and the executable program generation by reading the executable program code stored in the memory 701
The corresponding computer program of code, to realize following steps:
It obtains current session above, and candidate word recommended models is input to above by described, wherein, the candidate word pushes away
It is to be trained based on word dialog record as training sample using machine learning means to recommend model;
According to the candidate word recommended models, candidate word to be recommended is determined;
The candidate word to be recommended is shown in the candidate word display area of input interface.
The mobile terminal further includes:Peripheral Interface 703, RF (Radio Frequency, radio frequency) circuit 705, audio-frequency electric
Road 706, loud speaker 711, power management chip 708, input/output (I/O) subsystem 709, touch screen 712, other input/controls
Control equipment 710 and outside port 704, these components are communicated by one or more communication bus or signal wire 707.
It should be understood that diagram mobile terminal 700 is only an example of mobile terminal, and mobile terminal 700
Can have than more or less components shown in figure, two or more components can be combined or can be with
It is configured with different components.Various parts shown in figure can be including one or more signal processings and/or special
Hardware, software including integrated circuit are realized in the combination of hardware and software.
It is just provided in this embodiment below to be described in detail for the recommendation mobile terminal of candidate word, the mobile terminal
By taking mobile phone as an example.
Memory 701, the memory 701 can be by access such as CPU702, Peripheral Interfaces 703, and the memory 701 can
To include high-speed random access memory, nonvolatile memory can also be included, such as one or more disk memory,
Flush memory device or other volatile solid-state parts.
The peripheral hardware that outputs and inputs of equipment can be connected to CPU702 and deposited by Peripheral Interface 703, the Peripheral Interface 703
Reservoir 701.
I/O subsystems 709, the I/O subsystems 709 can be by the input/output peripherals in equipment, such as touch screen 712
With other input/control devicess 710, it is connected to Peripheral Interface 703.I/O subsystems 709 can include 7091 He of display controller
For controlling one or more input controllers 7092 of other input/control devicess 710.Wherein, one or more input controls
Device 7092 processed receives electric signal from other input/control devicess 710 or sends electric signal to other input/control devicess 710,
Other input/control devicess 710 can include physical button (pressing button, rocker buttons etc.), dial, slide switch, behaviour
Vertical pole clicks idler wheel.What deserves to be explained is input controller 7092 can with it is following any one connect:Keyboard, infrared port,
The indicating equipment of USB interface and such as mouse.
Touch screen 712, the touch screen 712 are the input interface and output interface between customer mobile terminal and user,
Visual output is shown to user, visual output can include figure, text, icon, video etc..
Display controller 7091 in I/O subsystems 709 receives electric signal from touch screen 712 or is sent out to touch screen 712
Electric signals.Touch screen 712 detects the contact on touch screen, and the contact detected is converted to and shown by display controller 7091
The interaction of user interface object on touch screen 712, that is, realize human-computer interaction, the user interface being shown on touch screen 712
Icon that object can be the icon of running game, be networked to corresponding network etc..What deserves to be explained is equipment can also include light
Mouse, light mouse are the extensions for not showing the touch sensitive surface visually exported or the touch sensitive surface formed by touch screen.
RF circuits 705 are mainly used for establishing the communication of mobile phone and wireless network (i.e. network side), realize mobile phone and wireless network
The data receiver of network and transmission.Such as transmitting-receiving short message, Email etc..Specifically, RF circuits 705 receive and send RF letters
Number, RF signals are also referred to as electromagnetic signal, and RF circuits 705 convert electrical signals to electromagnetic signal or electromagnetic signal is converted to telecommunications
Number, and communicated by the electromagnetic signal with communication network and other equipment.RF circuits 705 can include performing
The known circuit of these functions includes but not limited to antenna system, RF transceivers, one or more amplifiers, tuner, one
A or multiple oscillators, digital signal processor, CODEC (COder-DECoder, coder) chipset, user identifier mould
Block (Subscriber Identity Module, SIM) etc..
Voicefrequency circuit 706 is mainly used for receiving audio data from Peripheral Interface 703, which is converted to telecommunications
Number, and the electric signal is sent to loud speaker 711.
Loud speaker 711 for the voice signal for receiving mobile phone from wireless network by RF circuits 705, is reduced to sound
And play the sound to user.
Power management chip 708, the hardware for being connected by CPU702, I/O subsystem and Peripheral Interface are powered
And power management.
Mobile terminal provided by the embodiments of the present application can realize the effect for the recommendation applicability for improving candidate word.
It is arbitrary that recommendation apparatus, storage medium and the mobile terminal of the candidate word provided in above-described embodiment can perform the application
The recommendation method for the candidate word that embodiment is provided has and performs the corresponding function module of this method and advantageous effect.Not upper
The technical detail of detailed description in embodiment is stated, reference can be made to the recommendation method for the candidate word that the application any embodiment is provided.
Note that it above are only the preferred embodiment of the application and institute's application technology principle.It will be appreciated by those skilled in the art that
The application is not limited to specific embodiment described here, can carry out for a person skilled in the art it is various it is apparent variation,
The protection domain readjusted and substituted without departing from the application.Therefore, although being carried out by above example to the application
It is described in further detail, but the application is not limited only to above example, in the case where not departing from the application design, also
It can include other more equivalent embodiments, and scope of the present application is determined by scope of the appended claims.
Claims (10)
1. a kind of recommendation method of candidate word, which is characterized in that including:
It obtains current session above, and candidate word recommended models is input to above by described, wherein, the candidate word recommends mould
Type is to be trained based on word dialog record as training sample using machine learning means;
According to the candidate word recommended models, candidate word to be recommended is determined;
The candidate word to be recommended is shown in the candidate word display area of input interface.
2. the recommendation method of candidate word according to claim 1, which is characterized in that according to the candidate word recommended models,
Determine candidate word to be recommended, including:
According to the candidate word recommended models, primary candidate word to be recommended is determined;
The button operation of user is obtained, according to the button operation and the correspondence of the candidate word to be recommended, is treated to described
Recommended candidate word is screened again.
3. the recommendation method of candidate word according to claim 1, which is characterized in that according to the candidate word recommended models,
Determine candidate word to be recommended, including:
According to the candidate word recommended models, primary candidate word to be recommended is determined;
Input content of the user in dialog box is obtained, according to the matching of the input content and the candidate word to be recommended
Relationship screens the candidate word to be recommended again.
4. according to the recommendation method of claim 1-3 any one of them candidate words, which is characterized in that further include:
It obtains the word dialog record that history generates and is used as training sample, using machine learning means, determine the word dialog
Foregoing description and the incidence relation of word hereinafter in record;
According to the incidence relation, word model hereinafter corresponding with foregoing description is determined, as candidate word recommended models.
5. according to the recommendation method of claim 1-3 any one of them candidate words, which is characterized in that before obtaining current session
Text, and it is input to the candidate word recommended models above by described, including:
Obtain the time of origin above of the current session;
By time of origin be in preset period of time described in be input to the candidate word recommended models above.
6. the recommendation method of candidate word according to claim 1, which is characterized in that in the candidate word show area of input interface
Before candidate word to be recommended described in domain views, further include:
Obtain the application environment of user's word input;
Corresponding nondeclarative memory dictionary is established according to the application environment;
When detecting that user opens destination application, nondeclarative memory dictionary corresponding with the destination application is determined;
In described program phrase library for temporary memory, according to the candidate word recommended models, candidate word to be recommended is determined.
7. the recommendation method of candidate word according to claim 1, which is characterized in that in the candidate word show area of input interface
Before candidate word to be recommended described in domain views, further include:
Dialogue is obtained above, according to the exchange way for determining user and conversation partner above;
Corresponding certain memory dictionary is established according to the exchange way;
When detecting the exchange way of user and current session other side there are during corresponding certain memory dictionary, in the specific note
Recall in dictionary, according to the candidate word recommended models, determine candidate word to be recommended.
8. a kind of recommendation apparatus of candidate word, which is characterized in that including:
MIM message input module for obtaining current session above, and is input to candidate word recommended models above by described,
In, the candidate word recommended models are to be trained based on word dialog record as training sample using machine learning means;
Candidate word determining module to be recommended, for according to the candidate word recommended models, determining candidate word to be recommended;
Candidate word recommending module, for showing the candidate word to be recommended in the candidate word display area of input interface.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The recommendation method of the candidate word as described in any in claim 1-7 is realized during row.
10. a kind of mobile terminal, including memory, processor and storage are on a memory and can be in the computer of processor operation
Program, which is characterized in that the processor realizes the time as described in any in claim 1-7 when performing the computer program
Select the recommendation method of word.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711391875.9A CN108132717A (en) | 2017-12-21 | 2017-12-21 | Recommendation method, apparatus, storage medium and the mobile terminal of candidate word |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711391875.9A CN108132717A (en) | 2017-12-21 | 2017-12-21 | Recommendation method, apparatus, storage medium and the mobile terminal of candidate word |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108132717A true CN108132717A (en) | 2018-06-08 |
Family
ID=62391097
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711391875.9A Pending CN108132717A (en) | 2017-12-21 | 2017-12-21 | Recommendation method, apparatus, storage medium and the mobile terminal of candidate word |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108132717A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111124150A (en) * | 2019-12-25 | 2020-05-08 | 王少白 | Input method capable of visually reflecting emotion |
CN111506718A (en) * | 2020-04-20 | 2020-08-07 | 腾讯科技(深圳)有限公司 | Session message determining method, device, computer equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101452453A (en) * | 2007-11-29 | 2009-06-10 | 北京搜狗科技发展有限公司 | Input method web site navigation method and input method system |
CN103927299A (en) * | 2014-04-25 | 2014-07-16 | 百度在线网络技术(北京)有限公司 | Method for providing candidate sentences in input method and method and device for recommending input content |
US20150142785A1 (en) * | 2013-11-18 | 2015-05-21 | Palo Alto Research Center Incorporated | Generalized graph, rule, and spatial structure based recommendation engine |
CN105095182A (en) * | 2014-05-22 | 2015-11-25 | 华为技术有限公司 | Reply information recommendation method and apparatus |
CN106896932A (en) * | 2016-06-07 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of candidate word recommends method and device |
-
2017
- 2017-12-21 CN CN201711391875.9A patent/CN108132717A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101452453A (en) * | 2007-11-29 | 2009-06-10 | 北京搜狗科技发展有限公司 | Input method web site navigation method and input method system |
US20150142785A1 (en) * | 2013-11-18 | 2015-05-21 | Palo Alto Research Center Incorporated | Generalized graph, rule, and spatial structure based recommendation engine |
CN103927299A (en) * | 2014-04-25 | 2014-07-16 | 百度在线网络技术(北京)有限公司 | Method for providing candidate sentences in input method and method and device for recommending input content |
CN105095182A (en) * | 2014-05-22 | 2015-11-25 | 华为技术有限公司 | Reply information recommendation method and apparatus |
CN106896932A (en) * | 2016-06-07 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of candidate word recommends method and device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111124150A (en) * | 2019-12-25 | 2020-05-08 | 王少白 | Input method capable of visually reflecting emotion |
CN111506718A (en) * | 2020-04-20 | 2020-08-07 | 腾讯科技(深圳)有限公司 | Session message determining method, device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108121799A (en) | Recommendation method, apparatus, storage medium and the mobile terminal of revert statement | |
US11302302B2 (en) | Method, apparatus, device and storage medium for switching voice role | |
US10379712B2 (en) | Conversation user interface | |
CN110334360B (en) | Machine translation method and device, electronic device and storage medium | |
KR101491843B1 (en) | Conversation based search system and search method | |
JP6492069B2 (en) | Environment-aware interaction policy and response generation | |
CN108134876A (en) | Dialog analysis method, apparatus, storage medium and mobile terminal | |
CN108021572B (en) | Reply information recommendation method and device | |
CN108108142A (en) | Voice information processing method, device, terminal device and storage medium | |
CN107995370B (en) | Call control method, device, storage medium and mobile terminal | |
US20180210872A1 (en) | Input System Having a Communication Model | |
CN110263131B (en) | Reply information generation method, device and storage medium | |
CN107678823A (en) | Control method, device and the storage medium and mobile terminal of mobile terminal application | |
US20230091262A1 (en) | Interface display method and apparatus of application, device, and medium | |
JP5834291B2 (en) | Voice recognition device, automatic response method, and automatic response program | |
CN114154459A (en) | Speech recognition text processing method and device, electronic equipment and storage medium | |
US11862178B2 (en) | Electronic device for supporting artificial intelligence agent services to talk to users | |
CN111831806A (en) | Semantic integrity determination method and device, electronic equipment and storage medium | |
CN108132717A (en) | Recommendation method, apparatus, storage medium and the mobile terminal of candidate word | |
CN111079438A (en) | Identity authentication method and device, electronic equipment and storage medium | |
WO2019196645A1 (en) | Conversational information processing method, device, mobile terminal, and storage medium | |
CN108921530B (en) | Information judgment method and device, storage medium and terminal | |
CN108089897A (en) | Startup management method, device, storage medium and the mobile terminal of application program | |
CN109948155B (en) | Multi-intention selection method and device and terminal equipment | |
JP2018181018A (en) | Conversation providing device, conversation providing method, and program |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180608 |
|
RJ01 | Rejection of invention patent application after publication |