CN106227510A - Method and device is recommended in application - Google Patents
Method and device is recommended in application Download PDFInfo
- Publication number
- CN106227510A CN106227510A CN201610530129.2A CN201610530129A CN106227510A CN 106227510 A CN106227510 A CN 106227510A CN 201610530129 A CN201610530129 A CN 201610530129A CN 106227510 A CN106227510 A CN 106227510A
- Authority
- CN
- China
- Prior art keywords
- time quantum
- information
- recommended
- attribute information
- same time
- 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
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
Abstract
The present embodiments relate to a kind of application and recommend method and device, described method includes: receives and unlocks instruction, obtains described the first attribute information unlocking instruction;According to user behavior custom model and the first attribute information of described unblock instruction, determine object to be recommended;Show described object to be recommended.Described device includes: receiver module, is used for receiving unblock instruction, obtains described the first attribute information unlocking instruction;Recommending module, according to user behavior custom model and the first attribute information of described unblock instruction, determines object to be recommended;And present module, for showing described object to be recommended.Embodiment of the present invention behavioural habits based on user are to have the governed feature of rule, the user behavior that the behavioural habits opening application according to user are set up is accustomed to model and unlocks the first attribute information of instruction, determine and show object to be recommended, user can be facilitated to find the object wanting to open quickly, bring the biggest convenience and more preferable Consumer's Experience.
Description
Technical field
The present invention relates to terminal applies management technique field, particularly relate to a kind of application and recommend method and device.
Background technology
Along with software and hardware level is more and more higher, can with the quantity of the application (APP) of installation and operation the most increasingly in terminal
Huge.But, owing to being limited to the size in the top table face of terminal, the number of applications can listed in top table face is the most suitable
Limited, pacified to this end, existing terminal all allows the multiple desktop slidably switched of user setup to arrange out in terminal respectively
Each application of dress.
Normally, using for convenience, the most the most frequently used application can be placed on top table face (by terminal unlocking by user
First desktop of rear entrance is defined as top table face) so that the most frequently used application can be quickly found.But, a side
Face, along with terminal and the functions expanding of application, the terminal degree of depth enters in the middle of the Working Life that user is daily, and this also makes
The application that user often uses in terminal also gets more and more, and thus also begins to occur in that and is still difficult to arrange on a top table face
Go out the situation of the most conventional application.To this end, the solution that this is proposed by some manufacturers is to provide application searches function, Yong Huke
Quickly to be positioned application by function of search and then to open, desktop of leafing through page by page compared with user is found application and is wanted efficiently.
During realizing the present invention, inventor finds that prior art at least there is problems in that for a user, needs
Searched page to be entered also inputs Apply Names, and this is obvious or very inconvenient, and a lot of users desktop that would rather go to leaf through page by page is looked for
Application does not the most use the function of this application searches;And, on the other hand, even the conventional application listed on top table face
Or quickly navigated to the application wanting to use by search, system still can not run this application on one's own initiative, and needs etc. are used
After family unlocks the operation interface entering terminal, find the application wanting to open and perform opening operation, this application could be entered right
The runnable interface answered, and the operation that current a lot of application has waiting time longer initializes interface, this allows for user
Need just can see the main operation interface of application after waiting for a long time.
Summary of the invention
Embodiment of the present invention technical problem to be solved is, it is provided that method and device is recommended in a kind of application, can basis
After the behavioural habits recommendation terminal unlocking that user is daily, user wants the application opened.
For solving above-mentioned technical problem, the embodiment of the present invention provides following technical scheme: one applies recommendation method, bag
Include:
Receive and unlock instruction, obtain described the first attribute information unlocking instruction;
According to user behavior custom model and the first attribute information of described unblock instruction, determine object to be recommended;
Show described object to be recommended.
Further, the step setting up described user behavior custom model also includes:
In response to the triggering command to object, record the second attribute information of described object, and by the second attribute of described object
Information is associated with described object and preserves;
The second attribute information according to described object, obtains at the object of same time quantum internal trigger and described in the same time
Second attribute information of the object of unit internal trigger;
According to described second attribute information, the object of described same time quantum internal trigger is ranked up.
Further, described second attribute information includes: triggered time information, operation duration information and/or position letter
Breath;
Described the second attribute information according to described object, obtains the object in same time quantum internal trigger, including:
Triggered time information according to described object, mates described object with time quantum, obtains the same described time
The object that unit is corresponding.
Further, described according to described second attribute information, the object in described same time quantum is ranked up,
Including:
Object according to described same time quantum open number of times, the object in described same time quantum is ranked up;
If at least two object in described same time quantum to open number of times identical, then according to described at least two object
Run duration information, at least two object opening number of times identical described in described same time quantum is ranked up;
And/or,
Different weighted values is given respectively, according to described power by triggered time information, operation duration information and/or positional information
Second attribute information of the object in weight values and described same time quantum, calculates adding of the object in described same time quantum
Power mark;
The weighted score of the object according to described same time quantum, is ranked up the object in described same time quantum.
Further, described the first attribute information unlocking instruction includes: unlocked time information;
Described the first attribute information instructed according to user behavior custom model and described unblock, determines object to be recommended, bag
Include: time quantum residing when determining unblock according to described unlocked time information, described in described user behavior custom model
The object of the predetermined quantity that time quantum internal sort is forward is defined as object to be recommended.
Further, described method also includes:
Obtain the object to be recommended that user accepts, determine that the object to be recommended that described user accepts is accustomed to model pair at user behavior
The time quantum answered;
Improve in described time quantum, the sequence of the object to be recommended that described user accepts.
Further, described method also includes:
According to the object to be recommended determined, in advance at object to be recommended described in running background.
On the other hand, the present invention also provides for a kind of application recommendation apparatus, including:
Receiver module, is used for receiving unblock instruction, obtains described the first attribute information unlocking instruction;
Recommending module, according to user behavior custom model and the first attribute information of described unblock instruction, it is to be recommended right to determine
As;
Display module, for showing described object to be recommended.
Further, described device also includes:
Relating module, in response to the triggering command to object, records the second attribute information of described object, and by described object
Second attribute information is associated with described object and preserves;
Acquisition module, according to the second attribute information of described object, obtains at the object of same time quantum internal trigger and described
The second attribute information at same time quantum internal trigger object;
Order module, according to described second attribute information, is ranked up the object in described same time quantum.
Further, described second attribute information includes: triggered time information, operation duration information and/or position letter
Breath;
Described acquisition module includes: matching unit, according to the triggered time information of described object, by described object and time quantum
Mate, obtain the object that same described time quantum is corresponding.
Further, described order module includes:
First sequencing unit, opens number of times, in described same time quantum according to the object of described same time quantum
Object is ranked up;
Second sequencing unit, if at least two object in described same time quantum to open number of times identical, then according to described
The operation duration information of at least two object, to opening at least two object that number of times is identical described in described same time quantum
Application is ranked up;
And/or,
Weighted scoring unit, gives different weights respectively by triggered time information, operation duration information and/or positional information
Value, according to the second attribute information of the object in described weighted value and described same time quantum, calculates the described same time single
The weighted score of the object in unit;
3rd sequencing unit, according to the mark of the object of described same time quantum, to the object in described same time quantum
It is ranked up.
Further, described the first attribute information unlocking instruction includes: unlocked time information;
The time quantum that described recommending module is residing when determining unblock according to described unlocked time information, practises described user behavior
Described in used model, the object of the predetermined quantity that time quantum internal sort is forward is defined as object to be recommended.
Further, also include:
User accepts data obtaining module, obtains the object to be recommended that user accepts, and determine that described user accepts is to be recommended right
As the time quantum corresponding at user behavior custom model;
Put forward sequence module, improve in described time quantum, the sequence of the object to be recommended that described user accepts.
Further, described device also includes:
Prerun module, according to the object to be recommended determined, in advance at object to be recommended described in running background.
After using technique scheme, the present invention at least has the advantages that the embodiment of the present invention is based on user
Behavioural habits are to have the governed feature of rule, the user behavior custom model set up according to the behavior opening application of user and unblock
First attribute information of instruction, determines and shows object to be recommended, user can be facilitated to find the object wanting to open quickly,
Bring the biggest convenience and more preferable Consumer's Experience.Especially, can in a specific environment screen-unblocking backstage high precision
The application that user may want most to use is recommended in degree ground, therefrom selects application to be opened for user, brings greatly to user
Convenience and well experience.By the method and apparatus of the embodiment of the present invention, terminal can be allowed more intelligent, more in depth
Solve user.
Accompanying drawing explanation
Fig. 1 is the steps flow chart schematic diagram of embodiment of the present invention application recommendation method.
Fig. 2 is the steps flow chart schematic diagram setting up user behavior custom model of embodiment of the present invention application recommendation method.
Fig. 3 is the step of specific embodiment when setting up user behavior custom model of present invention application recommendation method
Schematic flow sheet.
Fig. 4 is the steps flow chart schematic diagram of another embodiment of present invention application recommendation method.
Fig. 5 is the steps flow chart schematic diagram of the another embodiment of present invention application recommendation method.
Fig. 6 is the functional-block diagram of embodiment of the present invention application recommendation apparatus.
Fig. 7 is the functional-block diagram of an embodiment of present invention application recommendation apparatus.
Fig. 8 is the functional-block diagram of the acquisition module of embodiment of the present invention application recommendation apparatus.
Fig. 9 is the functional-block diagram of the order module of embodiment of the present invention application recommendation apparatus.
Figure 10 is the functional-block diagram of another embodiment of present invention application recommendation apparatus.
Figure 11 is the functional-block diagram of another embodiment of present invention application recommendation apparatus.
Detailed description of the invention
With specific embodiment, the application is described in further detail below in conjunction with the accompanying drawings.Should be appreciated that following signal
Property embodiment and explanation only be used for explain the present invention, not as a limitation of the invention, and, in the case of not conflicting,
Embodiment in the application and the feature in embodiment can be combined with each other.
As it is shown in figure 1, the embodiment of the present invention provides a kind of object recommendation method, including:
S11, receives and unlocks instruction, obtains described the first attribute information unlocking instruction;
S12, according to user behavior custom model and the first attribute information of described unblock instruction, determines object to be recommended;
S13, shows described object to be recommended.
In above steps, described first attribute information that gets in step S11 will process as subsequent step
One of important evidence.And in step s 12, described object refers to the various application installed in terminal, it can be unit operation
Application, it is also possible to be the application needing networking operation.In step s 13, it is common that will eventually determine that object to be recommended shows
In the first interface entered after terminal unlocking, thus user can be quickly carried out selecting.
Based on user terminal use application behavioural habits substantially have rule governed, the present embodiment be i.e. according to
Family is usually opened the behavioural habits of application and is set up user behavior custom model, and combines the custom model reconciliation of described user behavior
Lock instruction relevant parameter information is comprehensively analyzed, and then determines application to be recommended, and application the most to be recommended presents
Select to user, such that it is able to allow user open the application wanting to use rapidly, hence it is evident that to bringing great convenience property of user
With good Consumer's Experience.Normally, the application to be recommended determined can be set as 3, certainly, will be set to other quantity also
It is feasible.Embodiment of the present invention method can be widely applied to intelligent terminal, such as: smart mobile phone, panel computer, Intelligent electric
Depending on etc., the application that may wish to run to facilitate user to be quickly found out when unlocking and entering.
Embodiment 2
On the basis of embodiment 1 more than the present invention, for setting up user behavior custom model, as in figure 2 it is shown, of the present invention
Method also includes:
S21, in response to the triggering command to object, records the second attribute information of described object, and by the second of described object
Attribute information is associated with described object and preserves;
S22, according to the second attribute information of described object, obtains at the object of same time quantum internal trigger and described same
Second attribute information of the object of time quantum internal trigger;
S23, according to described second attribute information, the object in triggering described same time quantum is ranked up.
In the step s 21, described association can be realized by the form setting up tables of data, each object in tables of data
Field under list the second attribute information that described object is corresponding, and described tables of data is saved in the memory module of terminal standby
With.
In the present embodiment, described time quantum is according to predetermined time span, divides a scoring cycle
And obtain, for the Elementary Time Unit analyzed and processed, in the specific implementation, for dividing the predetermined time of time quantum
Length can be identical, i.e. divides equally the scoring cycle, and in some optional embodiments of the present embodiment, with one day, (24 is little
Time) as a scoring cycle, and using 30 minutes as scheduled time length, is divided into 48 times single the described scoring cycle
Unit, each time quantum is 30 minutes.It is, of course, also possible to the scheduled time length of time quantum is set as that other time is long
Degree, such as;10 minutes, 20 minutes, 40 minutes, 60 minutes etc., thus obtain the time quantum of respective amount.Certainly, described pre-
Fixed time span also can also is that and differs, i.e. the scoring cycle does not divides equally, and can make according to user's application at ordinary times
Situation carry out classifying rationally, such as: by day in the period (as a example by 6:00 AM at 6 in evening), can with 30 minutes make
Dividing for scheduled time length, morning 0 every day then can be using 120 minutes as scheduled time length to 6:00 AM
Dividing, 6 then divided with 60 minutes for scheduled time length at 12 in evening in the afternoon.Scheduled time length
Treat with a certain discrimination, can preferably embody user's use habit.
In one embodiment, described second attribute information includes: triggered time information, operation duration information and/or position
Confidence ceases, and accordingly, in step S22, according to the second attribute information of described object, obtains in same time quantum internal trigger
Object include the most again:
Triggered time information according to described object, mates described object with time quantum, obtains the same described time
The object that unit is corresponding.
In the present embodiment, described triggered time information the most also may particularly include the moon, day, time, such as: on April 11 15:
00.Described second attribute information is typically to be carried out recording when the daily actually used each application of user by terminal and obtain and gradually
Collect and perfect, also can also be inputted by user voluntarily by arranging and obtain.
And in one embodiment, as it is shown on figure 3, above-mentioned step S23 can farther include the most again:
S231, opens number of times according to the object of described same time quantum, carries out the object in described same time quantum
Sequence;
S232, if at least two object in described same time quantum to open number of times identical, then according to described at least two
The operation duration information of object, arranges at least two object opening number of times identical described in described same time quantum
Sequence;
And/or,
S233, gives different weighted values, according to institute respectively by triggered time information, operation duration information and/or positional information
State the second attribute information of object in weighted value and described same time quantum, calculate the object in described same time quantum
Weighted score;
S234, according to the weighted score of the object of described same time quantum, is carried out the object in described same time quantum
Sequence.
In one embodiment, first it is to open number of times according to object in same time quantum to be ranked up, permissible
Effectively the application opened often is sorted front, and conveniently recommend.And if there is also at least two, to open number of times identical
During object, the most also can open the operation duration information of the identical object of number of times according to described at least two further and be ranked up,
It is typically the object order by running duration length front.
And in another embodiment, by triggered time information, operation duration information and/or positional information respectively
Give different weighted values, thus according to the second attribute information of the object in described weighted value and described same time quantum,
Calculate the weighted score of the object obtained in described same time quantum;And then further according to the object of described same time quantum
Object in described same time quantum is ranked up by weighted score.
In view of user's complexity to each application use habit, each application gathered when setting up user behavior custom model
Application operation information for applying the power of influence size of final ranking results all it may happen that change in each time quantum, because of
And, multiple sort algorithm can be used simultaneously to come simultaneously each application theed open in time quantum is ranked up, different rows
In sequence algorithm, the types of applications operation information weighted value when being ranked up application, power of influence is by different in other words, tool
Body is presented as on weighted value during application sequence.When processing using multiple sort algorithms to be ranked up, it is also possible to the most each
The ranking results of sort algorithm recommends application, such as: sort each sort algorithm in the application of former positions (such as: front three)
Carry out again collecting statistics, from which further follow that the recommended number of times of each application, then take recommended number of times or gross score comes former
To present to user for you to choose for the formal recommendation that should be used for of position (such as: front three);Or, also can be when sequence, according to each
Each application is marked by the weight of item application operation information, draws each sort algorithm marking and queuing result, then collects system
Count the gross score of application of former positions (such as: front three) that each sort algorithm is recommended, then take gross score and rank in the top (example
Such as front three) the formal recommendation that should be used for present to user for you to choose.
In actual use, then by constantly increase newly new sort algorithm participate in application sequence or to existing
Sort algorithm carries out suitable correction, thus, use the raising of the number of times of application along with user, by answering under different sort algorithms
Finally select the application used to compare with sort recommendations result and user, determine actual similarity, and preset a phase
Seemingly spending threshold value, when actual similarity is more than or equal to this similarity threshold, i.e. retain this sort algorithm, otherwise, i.e. eliminating should
Sort algorithm.Thus, the present embodiment can continue to optimize the sort algorithm of reservation, eliminates and recommends application finally to select to make with user
The relatively low sort algorithm of the similarity of application.Owing to sort algorithm itself constantly can enter along with the service condition of user
Chemistry is practised, and can effectively promote the accuracy of the sort algorithm remained.Reach the effect of algorithm optimization.
Embodiment 3
Based on the various embodiments described above, specific in the present embodiment 3, described the first attribute information unlocking instruction includes: during unblock
Between information;
Described step S12 specifically includes: time quantum residing when determining unblock according to described unlocked time information, by described use
Described in the behavioural habits model of family, the object of the predetermined quantity that time quantum internal sort is forward is defined as object to be recommended.
May determine that the time quantum residing for unblock by unlocked time, carrying out can be from the user behavior established in advance
Custom module obtains the object of the forward predetermined quantity of described time quantum internal sort be used as object to be recommended and present to
User, especially it is possible to first interface entered after terminal unlocking manifests the described object waiting recommendation, it is thus possible to convenient
User selects the object wanting to open quickly.
Finding in actual implementation process, some technical dates are likely to produce large effect for ranking results,
Therefore, when sequence, it is also possible to first confirm whether the triggered time is technical dates.Typically refer to legal joint described technical dates false
Day, also can also is that the commemoration day (such as: birthday, wedding anniversary etc.) relevant to the specific people of user or user setup.
If gram determining that the triggered time is technical dates, when carrying out described sequence, need the object row that will in the past open in described technical dates
Sequence is front, such as: assume that the conventional data of terminal record show, application one is more to open in the legal festivals and holidays to make time flat
With, and applying two is more to open and use on weekdays time flat, then when current date is festivals or holidays, then by described application one row
Sequence is front.
Additionally, also can there are some differences in the object that user opens in different location, such as: application conventional at home and
Being on duty the conventional application in place can be different, such that it is able to according to user come in diverse location place the most accurately to
Family recommended.According to the positional information that terminal when unlocking is current, and current positional information is accustomed to model with user behavior
In positional information be compared to object is ranked up, it is common that current positional information and user behavior are accustomed to model
The object order that middle positional information matches is front.By considering the positional information impact on sequence, it is contemplated that when user unlocks
The frequency of positional information variation relatively low, therefore, can for obtaining the time interval of positional information when terminal is unlocked
Relatively long to be set to, such as 1 day, i.e. obtain a current location information every day, specifically, can be when every day fixing
Between point (such as 8:30 in morning) obtain user's current location information, and in 24 hours effectively.It is apparent that it is as more accurate in obtained
Positional information application is sorted the most accurately, also can also suitably shorten position when adjacent twice acquisition unlocks
The time interval of information, such as: foreshorten to be spaced every 12 hours, is spaced every 4 hours, the most shorter.
The application of the present invention recommends method directly to realize in terminal, also can also realize by means of server.
As realized by means of server, then it is responsible for gathering relevant application operation information by terminal and uploads onto the server and set up user's row
For being accustomed to model, and the relevant parameter information unlocking instruction combining the terminal of acquisition realizes application sequence and determines to be recommended
Application, the information of application the most to be recommended be presented in terminal for user select.
In another alternative embodiment of the present invention, as shown in Figure 4, described method also includes:
S14, obtains the object to be recommended that user accepts, determines that the object to be recommended that described user accepts is accustomed to mould at user behavior
The time quantum that type is corresponding;
S15, improves in described time quantum, the sequence of the object to be recommended that described user accepts.
By user behavior custom model in described time quantum in increase accept recommendation information, to the described time
When application during unit is ranked up, herein in connection with the described recommendation information that accepts, application is ranked up.According to each time quantum
Interior user selects the application opened, each application when updating described time quantum accept recommendation information.Described accept recommendation
Cease the part as user behavior custom model, when determining application to be recommended, also by determined by be recommended should
Result produce impact.Thus, when the same time quantum in next cycle determines object to be recommended, described use
The object that family accepts will obtain the most forward ranking, so that object to be recommended is to be recommended with what actually user accepted
Object is the most identical.
By from above-mentioned many-sided influence factor considering sequence, such that it is able to set up user behavior custom more accurately
Model, and the recommendation accuracy of the object to be recommended determined in finally also can making each time quantum is higher.
In another alternative embodiment of the present invention, as it is shown in figure 5, the inventive method also can farther include:
S16, according to the object to be recommended determined, in advance at object to be recommended described in running background.
By the process of this step, the application that user may select can be run in advance on backstage in advance, thus, when with
When family have selected described application, the runnable interface applied can be carried out more quickly, improve application service efficiency.
On the other hand, for preferably implementing said method, the embodiment of the present invention also provides for a kind of application recommendation apparatus, such as figure
Shown in 6, described device includes:
Receiver module 10, is used for receiving unblock instruction, obtains described the first parameter information unlocking instruction;
Recommending module 20, according to user behavior custom model and the first parameter information of described unblock instruction, determines to be recommended
Object;
Display module 30, for showing described object to be recommended.
The present invention is to use the behavioural habits of application substantially to have the governed feature of rule based on user in terminal, and can basis
User usually opens the behavioural habits of application and sets up user behavior custom model, and then can be accustomed to mould in conjunction with described user behavior
Type and unblock instruction relevant parameter information are comprehensively analyzed, and determine application to be recommended, and application the most to be recommended shows
Show and select to user.Thus, it is possible to allow user open the application wanting to use rapidly, hence it is evident that bring great convenience to user
Property and good Consumer's Experience.The inventive method can be widely applied to intelligent terminal, such as: smart mobile phone, panel computer, intelligence
Energy TV etc., the application that may wish to run to facilitate user to be quickly found out when unlocking and entering.
In another alternative embodiment of the present invention, for setting up user behavior custom model, as it is shown in fig. 7, the present invention is real
Execute device described in example also to include:
Relating module 40, in response to the triggering command to object, records the second attribute information of described object, and by described object
The second attribute information be associated with described object and preserve;
Acquisition module 50, according to the second attribute information of described object, obtains in the object of same time quantum internal trigger and institute
State the second attribute information at same time quantum internal trigger object;
Order module 60, according to described second attribute information, is ranked up the object in described same time quantum.
Recorded by relating module 40 and preserve each object and corresponding the second attribute information associated thereof, and contributing to
The object of same time quantum internal trigger is obtained according to the second attribute information of object at acquisition module 50, and then can be by
Object is ranked up by order module 60 according to the second attribute information at same time quantum internal trigger object, thus can have
Effect sets up user behavior custom model, to facilitate follow-up determination object to be recommended.
In a specific embodiment, described second attribute information includes: triggered time information, run duration information and/
Or positional information.And as shown in Figure 8, described acquisition module 50 includes: matching unit 500, according to the triggered time of described object
Information, mates described object with time quantum, obtains the object that same described time quantum is corresponding.
In the present embodiment, described triggered time information the most also may particularly include the moon, day, time, such as: on April 11 15:
00.Described second attribute information is typically to be carried out recording when the daily actually used each application of user by terminal and obtain and gradually
Collect and perfect, also can also be inputted by user voluntarily by arranging and obtain.
And in another alternative embodiment of the invention, as it is shown in figure 9, described order module 60 also can be wrapped further
Include:
First sequencing unit 601, opens number of times, in described same time quantum according to the object of described same time quantum
Object be ranked up;
Second sequencing unit 602, if at least two object in described same time quantum to open number of times identical, then according to institute
State the operation duration information of at least two object, open, to described in described same time quantum, at least two pair that number of times is identical
As application is ranked up;
And/or,
Weighted scoring unit 603, gives different power respectively by triggered time information, operation duration information and/or positional information
Weight values, according to the second attribute information of the object in described weighted value and described same time quantum, calculates the described same time
The weighted score of the object in unit;
3rd sequencing unit 604, according to the mark of the object of described same time quantum, right in described same time quantum
As being ranked up.
In one embodiment, the first sequencing unit 601 open number of times according to object in same time quantum to enter
Row sequence, effectively can sort the application opened often front, and conveniently recommend.And if there is also at least two and open
During the identical object of number of times, the most also opened identical right of number of times according to described at least two further by the second sequencing unit 602
The operation duration information of elephant is ranked up, it is common that will run the object order of duration length front.This processing mode can make
Sorting operation is simple and relative efficiency.
And in another embodiment, by weighted scoring unit 603 to triggered time information, run duration information and/
Or positional information gives different weighted values respectively, thus according to the object in described weighted value and described same time quantum
Second attribute information, calculates the weighted score of the object obtained in described same time quantum;And then again by the 3rd sequencing unit
Object in described same time quantum is ranked up by 604 according to the weighted score of the object of described same time quantum.This
Plant and be ranked up according to weight mark, more can reflect the power of influence size of various influence factor.
In view of user's complexity to each application use habit, each application gathered when setting up user behavior custom model
Application operation information for applying the power of influence size of final ranking results all it may happen that change in each time quantum, because of
And, multiple sort algorithm can be used simultaneously to come simultaneously each application theed open in time quantum is ranked up, in different sequences
In algorithm, the types of applications operation information weighted value when being ranked up application, power of influence is by different in other words, specifically
It is presented as on weighted value during application sequence.When processing using multiple sort algorithms to be ranked up, it is also possible to comprehensive each row
The ranking results of sequence algorithm recommend application, such as: by each sort algorithm sequence former positions (such as: front three) application again
Carry out collecting statistics, from which further follow that the recommended number of times of each application, then take recommended number of times or gross score ranks in the top
To present to user for you to choose for the formal recommendation that should be used for of (such as: front three);Or, also can be when sequence, according to each item
Each application is marked by the weight of application operation information, draws each sort algorithm marking and queuing result, then collects statistics
The gross score of application of former positions (such as: front three) that each sort algorithm is recommended, then take gross score rank in the top (such as:
Front three) the formal recommendation that should be used for present to user for you to choose.
And, in actual use, moreover it is possible to participate in the sequence of application by constantly increasing new sort algorithm newly
Or existing sort algorithm is carried out suitable correction, thus, use the raising of the number of times of application along with user, difference sequence is calculated
The application sort recommendations result of FAXIA and user finally select the application used to compare, and determine actual similarity, and set in advance
A fixed similarity threshold, when actual similarity is more than or equal to this similarity threshold, i.e. retains this sort algorithm, otherwise,
I.e. eliminate this sort algorithm.Thus, the present embodiment can continue to optimize the sort algorithm of reservation, eliminates and recommends application with user
Select the sort algorithm that the similarity of the application of use is relatively low eventually.Owing to sort algorithm itself can be along with the use feelings of user
The continuous evolutionary learning of condition, and can effectively promote the accuracy of the sort algorithm remained.Reach the effect of algorithm optimization.
In another embodiment of the present invention, the first attribute information of the described unblock instruction acquired in receiver module 10
Including: unlocked time information.And described recommending module 20 time residing when determining unblock according to described unlocked time information is single
Unit, is defined as to be recommended by the object of predetermined quantity forward for time quantum internal sort described in described user behavior custom model
Object.
Determining the time quantum residing for unblock by unlocked time information, carrying out can be from the user behavior established in advance
Custom module obtains the object of the forward predetermined quantity of described time quantum internal sort be used as object to be recommended and present to
User, especially it is possible to first interface entered after terminal unlocking manifests the described object waiting recommendation, it is thus possible to convenient
User selects the object wanting to open quickly.
Owing to technical dates produces considerable influence for ranking results, therefore, when sequence, the triggered time also can first be confirmed
Whether it it is technical dates.Described technical dates typically refers to the legal festivals and holidays, and also can also be with user or user setup is specific
The commemoration day (such as: birthday, wedding anniversary etc.) that personnel are correlated with, however, it is determined that the triggered time is technical dates, carries out described row
Need the conventional object order opened in described technical dates during sequence front, such as: assume that the conventional data of terminal record show
Showing, application one is more to open and use in the legal festivals and holidays time flat, and applying two is more to open on weekdays to make time flat
With, then when current date is festivals or holidays, then described application one is sorted front.
Additionally, also can there are some differences in the object that user opens in different location, such as: application conventional at home and
Being on duty the conventional application in place can be different, such that it is able to according to user come in diverse location place the most accurately to
Family recommended.According to the positional information that terminal when unlocking is current, and current positional information is accustomed to model with user behavior
In positional information be compared to object is ranked up, it is common that current positional information and user behavior are accustomed to model
The object order that middle positional information matches is front.By considering the positional information impact on sequence, it is contemplated that when user unlocks
The frequency of positional information variation relatively low, therefore, can for obtaining the time interval of positional information when terminal is unlocked
Relatively long to be set to, such as 1 day, i.e. obtain a current location information every day, specifically, can be when every day fixing
Between point (such as 8:30 in morning) obtain user's current location information, and in 24 hours effectively.It is apparent that it is as more accurate in obtained
Positional information application is sorted the most accurately, also can also suitably shorten position when adjacent twice acquisition unlocks
The time interval of information, such as: foreshorten to be spaced every 12 hours, is spaced every 4 hours, the most shorter.
As shown in Figure 10, in another alternative embodiment of the present invention, assembly of the invention also includes:
User accepts data obtaining module 70, obtains the object to be recommended that user accepts, determines that described user accepts to be recommended
Object is accustomed to, at user behavior, the time quantum that model is corresponding;
Put forward sequence module 80, improve in described time quantum, the sequence of the object to be recommended that described user accepts.
By user behavior custom model in described time quantum in increase accept recommendation information, to the described time
When application during unit is ranked up, herein in connection with the described recommendation information that accepts, application is ranked up.According to each time quantum
Interior user selects the application opened, each application when updating described time quantum accept recommendation information.Described accept recommendation
Cease the part as user behavior custom model, when determining application to be recommended, also by determined by be recommended should
Result produce impact.Thus, when the same time quantum in next cycle determines object to be recommended, described use
The object that family accepts will obtain the most forward ranking, so that object to be recommended is to be recommended with what actually user accepted
Object is the most identical.
By from above-mentioned many-sided influence factor considering sequence, such that it is able to set up user behavior custom more accurately
Model, and the recommendation accuracy of the object to be recommended determined in finally also can making each time quantum is higher.
In an alternate embodiment of the present invention where, as shown in figure 11, assembly of the invention also can farther include:
Prerun module 90, according to the object to be recommended determined, in advance at object to be recommended described in running background.
The specific implementation process of the present invention is described below by way of an example.
First, it is a scoring cycle with one day (24 hours), and using every 30 minutes as a time quantum by one day
It is divided into 48 time quantums, during user uses terminal, answering of each application opened in recording each time quantum
With operation information, including the temporal information of each application, the most specifically open the moon of described application, day, time (such as: on April 11 15:
30), running duration (such as: run 10 minutes), when opening application, the current location information of user is (such as: XX road, XX district of XX city
XX mansion/community).
When setting up user behavior custom model, the object in each time quantum is ranked up, in this citing,
Being to be ranked up according to weighted value, therefore, each sort algorithm can be according to triggered time information, operation duration information and/or position
The weighted value of information, to the marking and queuing of each application in time quantum belonging to current, specifically according to each application score from height to
Low order sorts, and the scoring event of each application in one example is as shown in the table:
Application 1 | Application 2 | Application 3 | Application 4 | … | |
00:00-00:30 | 80 | 10 | 30 | 35 | … |
… | … | … | … | … | … |
09:01-09:30 | 35 | 90 | 86 | 85 | … |
… | … | … | … | … | … |
Thus, according to the application operation information gathered, it is established that user behavior custom model uses from each link of postorder.
Secondly, when user is unlocked screen operation at certain time point to terminal, the first parameter unlocking instruction is obtained
Information, specifically includes that unlocked time information (such as: April 11 15:26), and the positional information that user is current.
Affiliated time quantum is determined according to unlocked time information, and then when obtaining described from user behavior custom model
Between the sequence of each application in unit, will sort forward (as;Front three) application be defined as application to be recommended, and finally exist
User is presented to for you to choose in terminal.
The terminal related in the present invention refers to the calculating equipment with corresponding operating system platform, such as: smart mobile phone,
Panel computer, computer (desktop computer, notebook computer), reader etc.;Application refers to need to provide by means of corresponding platform
The software used to user, such as, Android application, Fructus Mali pumilae application, windows application etc.;Server can be single network clothes
Business device, the server group of multiple webserver composition or based on cloud computing be made up of a large amount of computers or the webserver
Cloud.
If the function described in the embodiment of the present invention realizes and as independent using the form of software function module or unit
When production marketing or use, a calculating device-readable can be stored in and take in storage medium.Based on such understanding, the present invention
Part or the part of this technical scheme that prior art is contributed by embodiment can embody with the form of software product
Coming, this software product is stored in a storage medium, including some instructions with so that a calculating equipment (can be individual
Computer, server, mobile computing device or the network equipment etc.) perform the whole of method described in each embodiment of the present invention or
Part steps.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory),
Random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store program code
Medium.In this specification, each embodiment uses the mode gone forward one by one to describe, and what each embodiment stressed is real with other
Executing the difference of example, between each embodiment, same or similar part sees mutually.
Although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, permissible
Understand and these embodiments can be carried out multiple change without departing from the principles and spirit of the present invention, revise, replace
And modification, the scope of the present invention is limited by claims and equivalency range thereof.
Claims (14)
1. an object recommendation method, it is characterised in that including:
Receive and unlock instruction, obtain described the first attribute information unlocking instruction;
According to user behavior custom model and the first attribute information of described unblock instruction, determine object to be recommended;
Show described object to be recommended.
2. the method for claim 1, it is characterised in that the step setting up described user behavior custom model includes:
In response to the triggering command to object, record the second attribute information of described object, and by the second attribute of described object
Information is associated with described object and preserves;
The second attribute information according to described object, obtains at the object of same time quantum internal trigger and described in the same time
Second attribute information of the object of unit internal trigger;
According to described second attribute information, the object of described same time quantum internal trigger is ranked up.
3. method as claimed in claim 2, it is characterised in that described second attribute information includes: triggered time information, operation
Duration information and/or positional information;
Described the second attribute information according to described object, obtains the object in same time quantum internal trigger, including:
Triggered time information according to described object, mates described object with time quantum, obtains the same described time
The object that unit is corresponding.
4. method as claimed in claim 3, it is characterised in that described according to described second attribute information, to described with for the moment
Between object in unit be ranked up, including:
Object according to described same time quantum open number of times, the object in described same time quantum is ranked up;
If at least two object in described same time quantum to open number of times identical, then according to described at least two object
Run duration information, at least two object opening number of times identical described in described same time quantum is ranked up;
And/or,
Different weighted values is given respectively, according to described power by triggered time information, operation duration information and/or positional information
Second attribute information of the object in weight values and described same time quantum, calculates adding of the object in described same time quantum
Power mark;
The weighted score of the object according to described same time quantum, is ranked up the object in described same time quantum.
5. the method as described in any one of claim 2-4, it is characterised in that the first attribute information bag of described unblock instruction
Include: unlocked time information;
Described the first attribute information instructed according to user behavior custom model and described unblock, determines object to be recommended, bag
Include: time quantum residing when determining unblock according to described unlocked time information, described in described user behavior custom model
The object of the predetermined quantity that time quantum internal sort is forward is defined as object to be recommended.
6. method as claimed in claim 2, it is characterised in that described method also includes:
Obtain the object to be recommended that user accepts, determine that the object to be recommended that described user accepts is accustomed to model pair at user behavior
The time quantum answered;
Improve in described time quantum, the sequence of the object to be recommended that described user accepts.
7. the method for claim 1, it is characterised in that described method also includes:
According to the object to be recommended determined, in advance at object to be recommended described in running background.
8. an application recommendation apparatus, it is characterised in that including:
Receiver module, is used for receiving unblock instruction, obtains described the first attribute information unlocking instruction;
Recommending module, according to user behavior custom model and the first attribute information of described unblock instruction, it is to be recommended right to determine
As;
Display module, for showing described object to be recommended.
Apply recommendation apparatus the most as claimed in claim 8, it is characterised in that described device also includes:
Relating module, in response to the triggering command to object, records the second attribute information of described object, and by described object
Second attribute information is associated with described object and preserves;
Acquisition module, according to the second attribute information of described object, obtains at the object of same time quantum internal trigger and described
The second attribute information at same time quantum internal trigger object;
Order module, according to described second attribute information, is ranked up the object in described same time quantum.
Apply recommendation apparatus the most as claimed in claim 9, it is characterised in that described second attribute information includes: the triggered time
Information, operation duration information and/or positional information;
Described acquisition module includes: matching unit, according to the triggered time information of described object, by described object and time quantum
Mate, obtain the object that same described time quantum is corresponding.
11. apply recommendation apparatus as claimed in claim 10, it is characterised in that described order module includes:
First sequencing unit, opens number of times, in described same time quantum according to the object of described same time quantum
Object is ranked up;
Second sequencing unit, if at least two object in described same time quantum to open number of times identical, then according to described
The operation duration information of at least two object, to opening at least two object that number of times is identical described in described same time quantum
Application is ranked up;
And/or,
Weighted scoring unit, gives different weights respectively by triggered time information, operation duration information and/or positional information
Value, according to the second attribute information of the object in described weighted value and described same time quantum, calculates the described same time single
The weighted score of the object in unit;
3rd sequencing unit, according to the mark of the object of described same time quantum, to the object in described same time quantum
It is ranked up.
The 12. application recommendation apparatus as described in any one of claim 9-11, it is characterised in that the first genus of described unblock instruction
Property information includes: unlocked time information;
The time quantum that described recommending module is residing when determining unblock according to described unlocked time information, practises described user behavior
Described in used model, the object of the predetermined quantity that time quantum internal sort is forward is defined as object to be recommended.
13. apply recommendation apparatus as claimed in claim 9, it is characterised in that also include:
User accepts data obtaining module, obtains the object to be recommended that user accepts, and determine that described user accepts is to be recommended right
As the time quantum corresponding at user behavior custom model;
Put forward sequence module, improve in described time quantum, the sequence of the object to be recommended that described user accepts.
14. apply recommendation apparatus as claimed in claim 8, it is characterised in that described device also includes:
Prerun module, according to the object to be recommended determined, in advance at object to be recommended described in running background.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610530129.2A CN106227510A (en) | 2016-07-06 | 2016-07-06 | Method and device is recommended in application |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610530129.2A CN106227510A (en) | 2016-07-06 | 2016-07-06 | Method and device is recommended in application |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106227510A true CN106227510A (en) | 2016-12-14 |
Family
ID=57520019
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610530129.2A Pending CN106227510A (en) | 2016-07-06 | 2016-07-06 | Method and device is recommended in application |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106227510A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018227333A1 (en) * | 2017-06-12 | 2018-12-20 | 深圳市乃斯网络科技有限公司 | Positioning-based terminal app recommendation method and system |
CN109189282A (en) * | 2018-07-12 | 2019-01-11 | 维沃移动通信有限公司 | A kind of application recommended method, device and mobile terminal |
CN109241450A (en) * | 2018-10-30 | 2019-01-18 | 麒麟合盛网络技术股份有限公司 | The recommended method and device of screen locking content |
CN109684524A (en) * | 2018-12-14 | 2019-04-26 | 惠州Tcl移动通信有限公司 | Using recommended method, device, storage medium and electronic equipment |
CN110602322A (en) * | 2019-09-12 | 2019-12-20 | 北京车慧科技有限公司 | Method and device for layout of application icons and mobile terminal |
CN112328338A (en) * | 2020-09-15 | 2021-02-05 | 深圳Tcl新技术有限公司 | Application hiding method, device, equipment and computer readable storage medium |
CN116095230A (en) * | 2022-08-17 | 2023-05-09 | 荣耀终端有限公司 | Application program recommendation method and related device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103019737A (en) * | 2012-12-27 | 2013-04-03 | 北京小米科技有限责任公司 | Method and device for displaying shortcut of application program |
CN103309598A (en) * | 2012-03-15 | 2013-09-18 | 宇龙计算机通信科技(深圳)有限公司 | Touch screen unlocking method and communication terminal |
CN103336654A (en) * | 2013-03-27 | 2013-10-02 | 广东欧珀移动通信有限公司 | Icon adjusting method and icon adjusting device |
CN103365540A (en) * | 2013-06-27 | 2013-10-23 | 天津三星通信技术研究有限公司 | Method and device for managing display of page in portable terminal and terminal |
CN103984470A (en) * | 2014-05-29 | 2014-08-13 | 中国联合网络通信集团有限公司 | Icon display method and device |
CN104679536A (en) * | 2013-11-28 | 2015-06-03 | 索尼公司 | Method and device for starting application program and electronic device |
CN104915215A (en) * | 2015-06-30 | 2015-09-16 | 努比亚技术有限公司 | Icon recommending device and method |
CN105487853A (en) * | 2015-11-19 | 2016-04-13 | 广东小天才科技有限公司 | Method and system for creating application shortcut in screen-locking interface |
-
2016
- 2016-07-06 CN CN201610530129.2A patent/CN106227510A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103309598A (en) * | 2012-03-15 | 2013-09-18 | 宇龙计算机通信科技(深圳)有限公司 | Touch screen unlocking method and communication terminal |
CN103019737A (en) * | 2012-12-27 | 2013-04-03 | 北京小米科技有限责任公司 | Method and device for displaying shortcut of application program |
CN103336654A (en) * | 2013-03-27 | 2013-10-02 | 广东欧珀移动通信有限公司 | Icon adjusting method and icon adjusting device |
CN103365540A (en) * | 2013-06-27 | 2013-10-23 | 天津三星通信技术研究有限公司 | Method and device for managing display of page in portable terminal and terminal |
CN104679536A (en) * | 2013-11-28 | 2015-06-03 | 索尼公司 | Method and device for starting application program and electronic device |
CN103984470A (en) * | 2014-05-29 | 2014-08-13 | 中国联合网络通信集团有限公司 | Icon display method and device |
CN104915215A (en) * | 2015-06-30 | 2015-09-16 | 努比亚技术有限公司 | Icon recommending device and method |
CN105487853A (en) * | 2015-11-19 | 2016-04-13 | 广东小天才科技有限公司 | Method and system for creating application shortcut in screen-locking interface |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018227333A1 (en) * | 2017-06-12 | 2018-12-20 | 深圳市乃斯网络科技有限公司 | Positioning-based terminal app recommendation method and system |
CN109189282A (en) * | 2018-07-12 | 2019-01-11 | 维沃移动通信有限公司 | A kind of application recommended method, device and mobile terminal |
CN109241450A (en) * | 2018-10-30 | 2019-01-18 | 麒麟合盛网络技术股份有限公司 | The recommended method and device of screen locking content |
CN109684524A (en) * | 2018-12-14 | 2019-04-26 | 惠州Tcl移动通信有限公司 | Using recommended method, device, storage medium and electronic equipment |
CN110602322A (en) * | 2019-09-12 | 2019-12-20 | 北京车慧科技有限公司 | Method and device for layout of application icons and mobile terminal |
CN110602322B (en) * | 2019-09-12 | 2020-11-03 | 北京车慧科技有限公司 | Method and device for layout of application icons and mobile terminal |
CN112328338A (en) * | 2020-09-15 | 2021-02-05 | 深圳Tcl新技术有限公司 | Application hiding method, device, equipment and computer readable storage medium |
CN116095230A (en) * | 2022-08-17 | 2023-05-09 | 荣耀终端有限公司 | Application program recommendation method and related device |
CN116095230B (en) * | 2022-08-17 | 2023-10-20 | 荣耀终端有限公司 | Application program recommendation method, terminal device and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106227510A (en) | Method and device is recommended in application | |
CN106547887B (en) | Search recommendation method and device based on artificial intelligence | |
CN103593425B (en) | Preference-based intelligent retrieval method and system | |
CN101322125B (en) | Improving ranking results using multiple nested ranking | |
CN109255586B (en) | Online personalized recommendation method for e-government affairs handling | |
CN107526807A (en) | Information recommendation method and device | |
CN104090888A (en) | Method and device for analyzing user behavior data | |
JP6694447B2 (en) | Big data calculation method and system, program, and recording medium | |
CN104969221B (en) | Semi-structured data in formatted data base | |
WO2018128718A1 (en) | System and method of generating a semantic representation of a target image for an image processing operation | |
CN105069080A (en) | Document retrieval method and system | |
CN111737608B (en) | Method and device for ordering enterprise information retrieval results | |
CN102968419A (en) | Disambiguation method for interactive Internet entity name | |
CN111078997B (en) | Information recommendation method and device | |
CN112231593A (en) | Financial information intelligent recommendation system | |
CN105512156A (en) | Method and device for generation of click models | |
CN116308109A (en) | Enterprise policy intelligent recommendation and policy making system based on big data | |
Benavent et al. | FCA-based knowledge representation and local generalized linear models to address relevance and diversity in diverse social images | |
CN113761084B (en) | POI search ranking model training method, ranking device, method and medium | |
CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
Arevalillo-Herráez et al. | A relevance feedback CBIR algorithm based on fuzzy sets | |
CN112148942A (en) | Business index data classification method and device based on data clustering | |
US10621165B2 (en) | Need supporting means generating apparatus and method | |
CN113592589B (en) | Textile raw material recommendation method, device and processor | |
CN109902116A (en) | A kind of Eco-Design knowledge active pushing system and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161214 |