Embodiment 1
According to embodiments of the present invention, a kind of embodiment of the method for the sort method of application App neutron applications is additionally provided,
It should be noted that can be in the calculating of such as one group of computer executable instructions the step of the flow process of accompanying drawing is illustrated
Execute in machine system, and, although show logical order in flow charts, but in some cases, can be with
The order being different from herein executes shown or described step.
The embodiment of the method provided by the embodiment of the present application one can be in mobile terminal, terminal or similar fortune
Calculate in device and execute.As a example by running on computer terminals, Fig. 1 is in a kind of application App of the embodiment of the present invention
The hardware block diagram of the terminal of the sort method of son application.As shown in figure 1, terminal 10 can be wrapped
(processor 102 can include but is not limited to microprocessor to processor 102 to include one or more (only illustrating one in figure)
The processing meanss of device MCU or PLD FPGA etc.), for data storage memorizer 104, Yi Jiyong
Transmitting device 106 in communication function.It will appreciated by the skilled person that the structure shown in Fig. 1 is only to show
Meaning, which does not cause to limit to the structure of above-mentioned electronic installation.For example, terminal 10 may also include and compare Fig. 1
Shown in more or less component, or with the configuration different from shown in Fig. 1.
Memorizer 104 can be used for software program and the module for storing application software, the such as application in the embodiment of the present invention
Corresponding programmed instruction/the module of the sort method of App neutron applications, processor 102 are stored in memorizer 104 by operation
Interior software program and module, so as to execute various function application and data processing, that is, realize above-mentioned application journey
The leak detection method of sequence.Memorizer 104 may include high speed random access memory, may also include nonvolatile memory,
Such as one or more magnetic storage device, flash memory or other non-volatile solid state memories.In some instances,
Memorizer 104 can further include the memorizer remotely located relative to processor 102, and these remote memories can be with
By network connection to terminal 10.The example of above-mentioned network include but is not limited to the Internet, intranet,
LAN, mobile radio communication and combinations thereof.
Transmitting device 106 is used for receiving via a network or sends data.Above-mentioned network instantiation may include
The wireless network that the communication providerses of terminal 10 are provided.In an example, transmitting device 106 includes one
Network adapter (Network Interface Controller, NIC), its can pass through base station and other network equipments
It is connected so as to can be communicated with the Internet.In an example, transmitting device 106 can be radio frequency (Radio
Frequency, RF) module, which is used for wirelessly being communicated with the Internet.
Under above-mentioned running environment, this application provides the sort method of application App neutron applications as shown in Figure 2.
Fig. 2 is the flow chart of the sort method of according to embodiments of the present invention one application App neutron applications.
As shown in Fig. 2 the sort method of application App neutron applications can include step is implemented as described below:
Step S202, obtains the customer attribute information using the first application App, and wherein, customer attribute information is used for referring to
Show that the information of the feature of user, the first application App include multiple first son applications.
It is soft that the first application App in the application above-mentioned steps S202 is not limited to the applications such as Alipay, Taobao, Taobao's travelling
Part, the first son application are not limited to purchase the air ticket, pay the application such as water power coal.User can be noted on the first application App
Volume logon account, after each user logs in the first application App using logon account, can pass through operation first and apply
App generates the operation information of each the first son application.
In the embodiment of the present invention, customer attribute information can serve to indicate that the feature of user, and the customer attribute information includes
But it is not limited to the following combination of one or more:Occupation, age, sex, consumption type and consumption degree.Wherein,
Occupation, age, sex, consumption type and consumption degree can be that on the first application App, registration is logged according to user
Speculated obtained from the information being input into during account, or by the operation information of user and obtained, for example, if user
A often purchases the air ticket and position is often changed, then the occupation that can deduce user A is probably businessperson, and
For example, if user B often buys women's dress, then the sex that can deduce user B is probably Ms.
Step S204, the application feature database that customer attribute information is pre-build with each the first son application is mated,
According to customer attribute information and the matching result of application feature database, each the first son application is ranked up.
The application feature database that customer attribute information is pre-build with each the first son application is mated, is specifically included:
The information similarity of application feature database that customer attribute information with each first son application pre-build, information are calculated respectively
Similarity is the matching result of customer attribute information and each application feature database.If the attribute of coupling is more, information is similar
Degree is then bigger, shows more match.
For each the first son application, the historic user attribute information of each the first son application is utilized respectively, is built in advance
The application feature database of each the first son application is found, that is, extracts the crowd characteristic at present using the first son application, side by side
Go out to quantify and orientable attribute tags, as the application feature database of the first son application.If user meets certain
The feature of the first son application is more, then user is bigger using the probability of the sub- application.
Further, the information similarity of the application feature database that is applied with each first son according to customer attribute information, to each
First son application is ranked up.
Alternatively, when S202 obtains the customer attribute information using the first application App, can also include:Obtain the
The operation information of each first son application of the user of one application App in the first preset time period, operation information include
Operation behavior information of the user of the first application App in each the first son application.
In the embodiment of the present invention, operation information can include the user of the first application App in each the first son application
Operation behavior information, the operation behavior information include the following combination of one or more:Last time use time, point
Hit number of times and payment times.For example, the collator of application App neutron applications can be purchased the air ticket to user A
Number of times, the number of times for clicking on application of purchasing the air ticket and last time were recorded using the time of application of purchasing the air ticket.
For example, as a example by applying A, after any one logon account for succeeding in registration Successful login Alipay, can
With the function of each the first son application in Alipay, specifically, apply the collator of App neutron applications count
In the first preset time period, the operation information of each application that (for example, in 3 months) Alipay is generated is (including behaviour
Make behavioural information), for example, within these three moons in January, 2015 to March, user A is applied for " air ticket "
Last time use time be on March 28th, 2015, number of clicks is 67 times, and payment times are 9 times;" water
The last time use time that electric coal " is applied is on January 31st, 2015, and number of clicks is 5 times, and payment times are
1 time;The last time use time that " washing in a pan point point " applies is on January 12nd, 2015, and number of clicks is 16 times,
Payment times are 5 times, and apply the collator of App neutron applications obtain the customer attribute information of Alipay,
For example, the attribute information of user A includes:Sex man, 37 years old age, professional sales manager etc., apply App neutrons
The collator of application can be according to the above- mentioned information for acquiring, to " air ticket " application, " water power coal " application and " water
Electric coal " application is ranked up.
Then, the customer attribute information and each the first son application feature database for pre-building of application are carried out in S204
Matching somebody with somebody includes:According to operation behavior information, the weight of operation behavior information and customer attribute information, it is calculated respectively
The recommendation degree score of each the first son application.
In the application above-mentioned steps, the weight of operation behavior information can be stored in advance in application App for designer
Son application collator in, specific weight value can by operator, BI (Business Intelligence,
Business intelligence) and product designer together decide on.
Still as a example by applying A, getting within these three moons in January, 2015 to March, user A is for " machine
The last time use time that ticket " is applied is on March 28th, 2015, and number of clicks is 67 times, and payment times are 9
Secondary;The last time use time that " water power coal " is applied is on January 31st, 2015, and number of clicks is 5 times, pays
Number of times is 1 time;The last time use time that " washing in a pan point point " applies is on January 12nd, 2015, and number of clicks is
16 times, payment times are 5 times, and the attribute information of user A includes:Sex man, 37 years old age, occupation pin
Sell manager etc., apply App neutron applications collator can according to above-mentioned data and the weight of operation behavior information,
" air ticket " application, " water power coal " apply and " wash in a pan a point point " apply recommendation degree score is calculated respectively.
The sort method of the application App neutron applications of the embodiment of the present invention can also original subscriber default first between in section
Using the behavioural habits of each the first son application, number of clicks and payment times therefore can be paid close attention to, and when weakening
Between factor, only for number of clicks and payment times close in the case of do successively sequence, therefore payment can be preset
The weight of number of times>The weight of number of clicks>>The weight of last time use time.
It should be noted that the sort method of the application App neutron applications of the embodiment of the present invention can be according to different passes
Note degree arranges corresponding weight to operation behavior information, all should be within the protection domain of the embodiment of the present invention.
Alternatively, also include after S204:Step S206, according to each first son application ranking results, to
First application App recommends the first son application.
In the application above-mentioned steps S206, apply the collator of App neutron applications be calculated each respectively
After the recommendation degree score of the first son application, according to the recommendation degree score of each the first son application, to the first application App
Recommend the first son application.The collator of application App neutron applications can be, but not limited to by generate include each the
The application recommendation list of one son application, and then application recommendation list is recommended the mode of the first application App according to each
The recommendation degree score of the first son application, recommends the first son application to the first application App.
Still as a example by applying A, the collator of App neutron applications is applied according to " air ticket " application, " water power coal "
The recommendation degree score that application and " washing in a pan point point " are applied, generates comprising " air ticket " application, " washing in a pan point point " application, " water
The application recommendation list that electric coal " is applied, after Alipay more redaction, is pushed to user with the application recommendation list
A.For user A, its " air ticket " application often to be used is arranged in forward position, " the water that seldom uses
Electric coal " application is arranged in rearward position.
From the foregoing, it will be observed that the scheme provided by the above embodiments of the present application one, by according to attribute information and each first
The operation information of son application is ranked up to each the first son application, has reached the real behavior data convert by user
The essential requirement of user, is suitable for the purpose of the application of each user's use habit to user's recommendation, it is achieved thereby that increasing
Plus the adaptive technique effect of application software, and then prior art is solved using during application software, version is more
After new, user is difficult to find the first son application that is conventional or thinking, the technical problem for causing application software adaptability poor.
Alternatively, described the customer attribute information is carried out with each the first son application feature database for pre-building of application
Coupling includes:According to the operation behavior information, the weight of the operation behavior information and the customer attribute information,
The recommendation degree score of each first son application is calculated respectively.
In a kind of alternative that the above embodiments of the present application are provided, as shown in figure 3, above-mentioned according to the user property
Information and the matching result of application feature database, are ranked up to each the first son application described, can include:
S302, according to the recommendation degree score of each the first son application, is ranked up to each the first son application.
In the application above-mentioned steps S302, apply the collator of App neutron applications realize in the way of employing is given a mark
Sequence to each the first son application, also, apply the collator of App neutron applications believe based on operation behavior
Breath, the weight of operation behavior information and customer attribute information, give a mark to each the first son application, also, root
According to the recommendation degree score of each the first son application, each the first son application is ranked up.
Alternatively, above-mentioned steps S302, based on operation behavior information, the weight of operation behavior information and user property
Information, carrying out marking to each the first son application can be using being implemented as follows scheme:
Step S3022, according to operation behavior information and the weight of operation behavior information, calculates each the first son application
The first score value.
In the application above-mentioned steps S3022, apply the collator of App neutron applications on the one hand can be gone according to operation
For information and the weight of operation behavior information, the first score value of each the first son application is calculated, on the other hand can be with root
According to customer attribute information, the second score value of each the first son application is calculated.First, the embodiment of the present invention is to applying App
The collator of neutron application how according to operation behavior information and the weight of operation behavior information, calculate each first
First score value of son application is described in detail:
Alternatively, above-mentioned steps S3022, according to operation behavior information and the weight of operation behavior information, calculate each
First score value of individual first son application can adopt and scheme is implemented as follows:
S10, is normalized to operation behavior information.
In the application above-mentioned steps S10, due to operation behavior information (for example, last time use time, click time
Number and payment times) not in a dimension, therefore the collator of application App neutron applications first can be to behaviour
It is normalized as behavioural information.Wherein, normalization is a kind of dimensionless processing means, will have the table of dimension
Formula is reached, through conversion, nondimensional expression formula is turned to, is become scalar.
Wherein, operation behavior information includes the following combination of one or more:Last time use time, number of clicks
And payment times.
Alternatively, in the case where operation behavior information includes last time use time, above-mentioned steps S10, to behaviour
It is normalized as behavioural information, can includes:Obtain in the user behavior sample set of collection in advance corresponding to most
The maximum of first use time and minima afterwards;By formula Y1=(R-Rmax)/(Rmax-Rmin) calculate normalizing
Last time use time after change, wherein, Y1Represent that the last time use time after normalization, R represent last
First use time, RmaxRepresent the maximum corresponding to last time use time, RminRepresent corresponding to last
The minima of secondary use time.
In the case where operation behavior information includes number of clicks, above-mentioned steps S10 are returned to operation behavior information
One change is processed, and can be included:Obtain in the in advance user behavior sample set of collection corresponding to number of clicks maximum and
Minima;By formula Y2=(F-Fmin)/(Fmax-Fmin) calculate the number of clicks after normalization, wherein, Y2
Represent that the number of clicks after normalization, F represent number of clicks, FmaxRepresent the maximum corresponding to number of clicks, Fmin
Represent the minima corresponding to number of clicks.
In the case where operation behavior information includes payment times, above-mentioned steps S10 are returned to operation behavior information
One change is processed, and can be included:Obtain in the in advance user behavior sample set of collection corresponding to payment times maximum and
Minima;By formula Y3=(M-Mmin)/(Mmax-Mmin) payment times after normalization are calculated, wherein,
Y3Represent that the payment times after normalization, F represent payment times, MmaxThe maximum corresponding to payment times is represented,
MminRepresent the minima corresponding to payment times.
The embodiment of the present invention to number of clicks and can be propped up during being normalized to operation behavior information
Pay the extreme value in number of times and all do especially process, in sample set, for example, remove the numerical value of obvious exception, to avoid affecting
The effect of normalized.
S12, by formulaThe first score value of each the first son application is calculated, wherein, S1 represents the
One score value, YiRepresent that the operation behavior information after normalization, n represent the number of operation behavior information, XiRepresent operation row
Weight for information.
In the application above-mentioned steps S12, the collator of App neutron applications is applied to return operation behavior information
After one change is processed, formula can be passed throughCalculate the first score value of each the first son application.
For example, in the case where operation behavior information includes last time use time, number of clicks and payment times,
S1=X1×Y1+X2×Y2+X3×Y3, wherein, Y1Represent the last time use time after normalization, Y2Expression is returned
Number of clicks after one change, Y3Represent the payment times after normalization, X1When representing that last time set in advance is used
Between corresponding weight, X2Represent the corresponding weight of number of clicks set in advance, X3Represent payment time set in advance
The corresponding weight of number, X3>X2>>X1.
Step S3024, customer attribute information is mated with the application feature database for pre-building, calculate each first
Second score value of son application.
In the application above-mentioned steps S3024, the application feature database that pre-builds can be in advance from each application,
The crowd characteristic that applies using each at present is extracted jointly by product manager and/or service operation people, and is listed and can be quantified
With orientable attribute tags, it is that each application determines corresponding attribute tags, and then obtains the application for pre-building
Feature database.Below, the embodiment of the present invention to apply the collator of App neutron applications how by customer attribute information with
The application feature database for pre-building is mated, and the second score value for calculating each the first son application is described in detail:
Alternatively, above-mentioned steps S3024, customer attribute information is mated with the application feature database for pre-building,
Second score value for calculating each the first son application can be using being implemented as follows scheme:
S20, searches the application matched with customer attribute information in the application feature database for pre-building.
In the application above-mentioned steps S20, the collator of application App neutron applications can be special in the application for pre-building
Levy and in storehouse, search the application matched with customer attribute information, for example, apply the collator of App neutron applications can be with
According to the occupation in customer attribute information, age, sex, consumption type and consumption degree etc. with each application above-mentioned really
Fixed corresponding attribute tags are mated, and then find the application matched with customer attribute information.
For example, the attribute mark that " stock market " application and " travel abroad " are applied in the application feature database for pre-building
Signing includes businessperson, and the attribute information of user A includes:Sex man, 37 years old age, professional sales manager, its
In may determine that user A is belonging to businessperson by occupation, then application App neutron applications collator then may be used
With find from the application feature database for pre-building " stock market " that mate with the attribute information of user A apply and
" travel abroad " is applied.
S22, gives preset fraction to the application matched with customer attribute information.
In above-mentioned steps S22, the collator of App neutron applications is applied to look in the application feature database for pre-building
After looking for the application matched with customer attribute information, the application that customer attribute information can be matched gives default point
Number, wherein, preset fraction can also operator, BI and product designer together decide on.For example, App is applied
The collator of neutron application can be to finding the attribute information with user A from the application feature database for pre-building
" stock market " application and " travel abroad " application that matches somebody with somebody gives the preset fraction, and unmatched application is not given
The preset fraction.
S24, according to preset fraction, calculates the second score value of each the first son application.
In the application above-mentioned steps S24, the collator of App neutron applications is applied to above-mentioned and customer attribute information
After the application for matching gives preset fraction, the second score value of each the first son application can be calculated.
Step S3026, to each first son application first score value sue for peace with the second score value, obtain each first
The score value of the corresponding recommendation degree score of son application.
In the application above-mentioned steps S3026, the of each the first son application for being obtained based on above-mentioned steps S10 to S12
One score value, and each first sub second score value that applies that above-mentioned steps S20 to S24 is obtained, apply App neutrons
The collator of application is sued for peace to first score value and the second score value, and then it is corresponding to obtain each the first son application
The score value of recommendation degree score.
S304, will come top n the first son application and recommend the first application App, and wherein N is default positive integer;Or
First son application of the recommendation degree score more than predetermined threshold value is recommended the first application App by person.
In the application above-mentioned steps S304, the collator of App neutron applications is applied to calculate each the first son application
Score value after, each the first son application can be ranked up according to score value order from big to small, and before coming
The first application App is recommended in N number of first son application, or, first son of the recommendation degree score more than predetermined threshold value is applied
Recommend the first application App.For example, the collator of application App neutron applications is calculated for user A, " machine
The fraction that ticket " is applied is more than " overseas more than the fraction that " stock market " is applied more than the fraction that " washing in a pan point point " applies
The fraction that the fraction " water power coal " that trip " is applied is applied, and then the collator of App neutron applications is applied according to score value
Order from big to small, according to " air ticket " application, " washing in a pan point point " application, " stock market " application, " travel abroad "
The order that application, " water power coal " are applied, and top n the first son application or recommendation degree score will be come more than default threshold
First son application of value generates above-mentioned application recommendation list.
It should be noted that the embodiment of the present invention be exemplary explanation can be right according to score value order from big to small
Each the first son application is ranked up, it would however also be possible to employ from small to large etc., the present invention is to this for other modes, such as score value
It is not restricted.
In a kind of alternative that the above embodiments of the present application are provided, as shown in figure 4, above-mentioned steps S206, according to each
The recommendation degree score of individual first son application, before recommending the first son application to the first application App, application App neutrons should
Sort method can also include:
S402, determines that the first application App does not generate operation information and the first application App in the first preset time period
The second son application of operation information is generated in the second preset time period.
Above-mentioned steps S202 apply the collator of App neutron applications in the first Preset Time into step S206
Each the first son application for generating operation information in section is ranked up, alternatively, in the application above-mentioned steps S402,
The collator of application App neutron applications can also be lost in user to the silence that applies carries out application recommendation, i.e., first
Operation information is not generated in preset time period and the first application App generates operation information in the second preset time period
Second son application (for example, be not used in nearly 3 months, but used in nearly 1 year).
Still as a example by applying A, user A is not used in nearly three months, but used application is in nearly 1 year
" book keeping operation is originally " application, then the collator of application App neutron applications can find out " the note according to above-mentioned condition
Account book " is applied.
S404, gives default score value to the second son application.
In above-mentioned steps S404, the collator of App neutron applications is applied to determine the first application App pre- first
If not generating operation information in the time period and the first application App generating the of operation information in the second preset time period
After two son applications, default score value can be given to the second son application.Identical, the default score value can also be operations
Business, BI and product designer together decide on.
Still as a example by applying A, the collator of App neutron applications is applied to find out " the note according to above-mentioned condition
After account book " application, score value can be given to " book keeping operation this " application, then follow-up according to score value from big to
Little when being ranked up to application, should be to " air ticket " application, " wash in a pan point a point " application, " stock market " application, " border
Outer trip " application, " water power coal " application and " book keeping operation is originally " application are ranked up jointly, so as to generate comprising " air ticket "
Application, " washing in a pan point point " application, " stock market " application, " travel abroad " application, " water power coal " application and " note
The application recommendation list that account book " is applied.
Alternatively, according to the recommendation degree score of each the first son application, recommend the first son application to the first application App,
Including:According to score value order from big to small, recommend the first son application and the second son application to the first application App.
In a kind of alternative that the above embodiments of the present application are provided, above-mentioned steps S206, according to each the first son application
Recommendation degree score, to first application App recommend first son application before, apply App neutron applications sort method
Can also include:
S30, obtains the 3rd son application that need to be arranged in before each the first son application.
Wherein, according to the recommendation degree score of each the first son application, the first son application, bag are recommended to the first application App
Include:3rd son is applied and each the first son application is ranked up, and deleted sub with the 3rd in each the first son application
Application identical application;Recommend the 3rd son application and each the first son application to the first application App, wherein, each
Do not include in first son application and the 3rd son application identical application.
In above-mentioned steps S30, based on operator to the popularization demand of application or based on the higher application of some importances,
The sort method of the application App neutron applications of the embodiment of the present invention can be obtained with generating before applying recommendation list
Need to be arranged in the 3rd son application before each the first son application.
Still as a example by applying A, " Yuebao " application, " transferring accounts " application, " prepaid mobile phone recharging " application, " credit card
Refund " application etc. belongs to the application or the application that importance is higher of operator's needs popularization, and these applications need to arrange
In forward position (in general position is fixed), for each, the position of these applications is identical, then
The collator of application App neutron applications can obtain " Yuebao " application before application recommendation list is generated, " turn
The applications such as account " application, " prepaid mobile phone recharging " application, " credit card repayment " application, and then " Yuebao " is applied,
" transferring accounts " application, " prepaid mobile phone recharging " are applied, " credit card repayment " is applied, " air ticket " is applied, " washing in a pan point point " answers
It is ranked up with, " stock market " application, " travel abroad " application, " water power coal " application and " book keeping operation this " application,
It should be noted that be possible to come across above-mentioned 3rd son application identical application in each the first son application above-mentioned, that
, in sequence, should delete in each the first son application and identical application be applied with the 3rd son, and then should to first
Recommend the 3rd son application and each the first son application with App, wherein, do not include and the 3rd in each the first son application
Son application identical application.
In a kind of alternative that the above embodiments of the present application are provided, above-mentioned steps S206, according to each the first son application
Recommendation degree score, to first application App recommend first son application before, apply App neutron applications sort method
Can also include:
S40, obtains the 4th son application that need to be arranged in after each first son is applied.
Wherein, according to the recommendation degree score of each the first son application, the first son application, bag are recommended to the first application App
Include:To the 4th son application and each first son application be ranked up, and delete the 4th son application in each first son
Application identical application;Recommend the 4th son application and each the first son application, wherein, the 4th to the first application App
Do not include in son application and each the first son application identical application.
In above-mentioned steps S40, based on popular use habit, the row of the application App neutron applications of the embodiment of the present invention
Sequence method can also be obtained to be needed to be arranged in the 4th son application before each the first son application.
Still as a example by applying A, " love donation " application, " AA gatherings " application, " financing small tool " are applied, " are gone
" application etc. belongs to the application of the use habit for meeting masses, these applications can be arranged in rearward position, then
Application App neutron applications collator generate application recommendation list before can obtain " love donation " application,
The applications such as " AA gatherings " application, " financing small tool " application, " going " application, and then should to " love donation "
With, " AA gatherings " application, " financing small tool " application, " going " application, " air ticket " application, " wash in a pan point a point "
Application, " stock market " application, " travel abroad " application, " water power coal " application and " keeping accounts this " application are arranged
Sequence, it should be noted that be possible to come across above-mentioned 4th son application identical application in each the first son application above-mentioned,
So, in sequence, should delete in the 4th son application with each the first son application identical application, and then to first
Application App recommends the 4th son application and each the first son application, wherein, do not include in the 4th son application with each the
One son application identical application.
You need to add is that, for the sort method of the application App neutron applications of the embodiment of the present invention, if application App
The collator of neutron application cannot obtain the user of the customer attribute information of the first application App and the first application App
Operation information of each the first son application in the first preset time period, for example, the user be download for the first time this
One application App, and from being not used, then the collator of application App neutron applications can obtain acquiescence just
Beginning list of application, and it is pushed to the first application App.
With reference to Fig. 5, the overall plan of the application is carried out exemplary description:
Step A, Part I (highest priority):The fixed position of the 3rd son application.
In the application above-mentioned steps A, based on popularization demand or based on some importances higher of the operator to application
Application, the sort method of the application App neutron applications of the embodiment of the present invention can with generating before applying recommendation list,
Obtaining needs to be arranged in the 3rd son application before each the first son application.
As a example by applying A, " Yuebao " application, " transferring accounts " application, " prepaid mobile phone recharging " application, " credit card repayment "
Application etc. belongs to the higher application of the application that operator needs promote or importance, these apply need to be arranged in forward
Position (in general position is fixed), for each, these application positions be identical, then application
The collator of App neutron applications can obtain " Yuebao " application, " transferring accounts " before application recommendation list is generated
The applications such as application, " prepaid mobile phone recharging " application, " credit card repayment " application
Step B, Part II (priority is taken second place):Personalization preferences of each user to application.
In the application above-mentioned steps B, can be reached according to customer attribute information and real user behavior (i.e. operation information)
To the essential requirement by the real behavior data convert user of user, generate and be suitable for answering for each user's use habit
Purpose with recommendation list.
Step B1, based on operation information, calculates the score of each the first son application.
Wherein, operation information can include operation behavior information, the operation behavior information include following one or more
Combination:Last time use time, number of clicks and payment times.For example, the sequence of App neutron applications is applied
Number of times, the number of times for clicking on application of purchasing the air ticket and the last time that device can be purchased the air ticket to user A is using purchase
The time of air ticket application is recorded.
Step B2, based on customer attribute information, calculates the score of each the first son application.
Wherein, customer attribute information can serve to indicate that the feature of user, the customer attribute information including but not limited to
The lower combination of one or more:Occupation, age, sex, consumption type and consumption degree.Wherein, occupation, the age,
Sex, consumption type and consumption degree are input into when can register logon account according to user on the first application App
Speculated obtained from information, or by the operation information of user and obtained, for example, if user A often buys machine
Ticket and position is often changed, then the occupation that can deduce user A is probably businessperson, and for example, if user
B often buys women's dress, then the sex that can deduce user B is probably Ms.
Step B3, the score that is applied based on each are ranked up to each the first son application.
Wherein, the application is adopted based on RFM, and superposition customer attribute information is used with application matching degree mode, structure
Personalization preferences model of the family to application.Wherein, R represents that user clicks on the last time time of certain application, and F represents use
The number of times (i.e. number of clicks) of the application is clicked at family in the first preset time period, and M represents user in the first Preset Time
Payment times in section, the first preset time period can be preset as needed.Its Computing Principle is by each
User applies last time use time, number of clicks, payment times in the first preset time period to count each
Come, then normalized in respective dimension, in conjunction with the weighted value of 3 factors, calculate each user and each is applied
Preference score value, fraction is higher to represent that the probability that is used by a user of application is higher.And customer attribute information is mated with application
Degree, refers to the occupation of user itself, trip, using features such as scenes, as application potential user's recommendation foundation, if
The feature of user meets application feature database, then the user is just added to the score value of the application.
Step C, Part III (priority is minimum):The sequence of the 4th son application.
In above-mentioned steps C, based on popular use habit, the sequence of the application App neutron applications of the embodiment of the present invention
Method can need to be arranged in the 4th son before each the first son application, after application recommendation list is generated, to obtain
Application.
Step D, generates application recommendation list.
It should be noted that the priority of final first application App homepage applications can be:User is actively arranged>Strategy
Fixed bit (i.e. above-mentioned Part I)>Intelligent sequencing (i.e. above-mentioned Part II) based on user behavior and feature>Acquiescence
List ordering (i.e. above-mentioned Part III), and then the application recommendation list of the personalization for being suitable for user is generated according to priority.
The sort method of application App neutron applications provided in an embodiment of the present invention, the sequence of application are first to gather user
The use time of last time, number of clicks, the real user behavior of 3 dimensions of payment times, build RFM models,
The true use habit that original subscriber applies is gone back to each, and provide the use score of each application;Again based on user's
Customer attribute information, such as occupation, age-sex, trip and consumption online feature etc., with the application feature for pre-building
Storehouse is mated, and gives adaptable reserved portion;For each user, above two score is added, and obtains final
Score, and sort according to score size, obtain the individualized section application sequence for adapting to user's use habit.And adopt
The mode merged with " specifying the fixed position+individualized section application sequence+default sort of application " more rules priority,
Application is carried out to user to recommend while application recommendation list can be generated.And this application recommendation list, system can preserve
Beyond the clouds, after version change, the application sequence of user will not change with version updating, reduce user each
The worry of conventional application is found again.That is, the sort method of the application App neutron applications of the embodiment of the present invention is obtained
The application recommendation list for going out, can respect fully user's custom, reduce user and find conventional application path, so as to optimize
Consumer's Experience.
In embodiments of the present invention, apply App's using the customer attribute information and first for obtaining the first application App
The operation information of each first son application of the user in the first preset time period, wherein, customer attribute information is used for referring to
Show that the information of the feature of user, operation information include the user of the first application App in each the first son application
Operation behavior information;According to operation behavior information, the weight of operation behavior information and customer attribute information, count respectively
Calculate the recommendation degree score for obtaining each the first son application;According to the recommendation degree score of each the first son application, should to first
The mode for recommending the first son application with App, by the operation information meter according to attribute information and each the first son application
The recommendation degree score for obtaining each the first son application is calculated, the sheet of the real behavior data convert user by user has been reached
Matter demand, is suitable for the purpose of the application of each user's use habit to user's recommendation, it is achieved thereby that it is soft to increase application
The adaptive technique effect of part, and then prior art is solved using during application software, user after version updating
It is difficult to find the first son application that is conventional or thinking, the technical problem for causing application software adaptability poor.
It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore which is all expressed as one it is
The combination of actions of row, but those skilled in the art should know, and the present invention is not limited by described sequence of movement
System, because according to the present invention, some steps can be carried out using other orders or simultaneously.Secondly, art technology
Personnel should also know that embodiment described in this description belongs to preferred embodiment, involved action and module
Not necessarily the present invention is necessary.
Through the above description of the embodiments, those skilled in the art is can be understood that according to above-mentioned enforcement
The method of example can add the mode of required general hardware platform by software to realize, naturally it is also possible to by hardware, but
The former is more preferably embodiment in many cases.Be based on such understanding, technical scheme substantially or
Say that the part contributed by prior art can be embodied in the form of software product, the computer software product is deposited
Storage is used so that a station terminal including some instructions in a storage medium (such as ROM/RAM, magnetic disc, CD)
Equipment (can be mobile phone, computer, server, or network equipment etc.) is executed described in each embodiment of the invention
Method.
Embodiment 2
According to embodiments of the present invention, a kind of device embodiment for implementing said method embodiment, this Shen are additionally provided
Please the device that provided of above-described embodiment can run on computer terminals.
Fig. 6 is the structural representation of the collator of the application App neutron applications according to the embodiment of the present application two.
As shown in fig. 6, the collator of application App neutron applications can include first acquisition unit 602, process
Unit 604 and sequencing unit 606.
Wherein, first acquisition unit 602, for obtaining the customer attribute information using the first application App, wherein,
The customer attribute information is used for the information of the feature of instruction user, and the first application App includes multiple first sons
Application;Processing unit 604, for the application spy for pre-building the customer attribute information with each the first son application
Levy storehouse to be mated;Sequencing unit 606, for the matching result according to the customer attribute information and application feature database,
Each the first son application described is ranked up.
From the foregoing, it will be observed that the scheme provided by the above embodiments of the present application two, by customer attribute information and application feature database
Matching result, to described each first son application be ranked up, reached the real behavior data convert by user
The essential requirement of user, and then the purpose of the application that can recommend to be suitable for each user's use habit to user, so as to
Achieve the increase adaptive technique effect of application software, and then prior art is solved using during application software,
After version updating, user is difficult to find the first son application that is conventional or thinking, the technology for causing application software adaptability poor
Problem.
Herein it should be noted that above-mentioned first acquisition unit 602, processing unit 604 and sequencing unit 606 pairs
Step S202 that should be in embodiment one to step S204, example that three modules are realized with corresponding step and should
Identical with scene, but it is not limited to one disclosure of that of above-described embodiment.It should be noted that above-mentioned module is used as dress
The part that puts is may operate in the terminal 10 of the offer of embodiment one, can be realized by software, it is also possible to
Realized by hardware.
Alternatively, processing unit 604 is used for executing following steps by the customer attribute information and each the first son application
The application feature database for pre-building is mated:The customer attribute information is calculated respectively with each the first son application in advance
The information similarity of the application feature database of foundation, described information similarity are that the customer attribute information is special with each application
Levy the matching result in storehouse.
Alternatively, described device also includes:Second acquisition unit, exists for obtaining the user of the first application App
The operation information of each the first son application in the first preset time period, wherein, the operation information includes described first
Operation behavior information of the user of application App in each the first son application.
Alternatively, processing unit 604 is used for executing following steps by the customer attribute information and each the first son application
The application feature database for pre-building is mated:According to the operation behavior information, the weight of the operation behavior information
And the customer attribute information, it is calculated the recommendation degree score of each the first son application respectively.
Alternatively, 606 sequencing unit of the sequencing unit be used for execute following steps according to the customer attribute information with
The matching result of application feature database, is ranked up to each the first son application described:According to each the first son application described
Recommendation degree score, to described each first son application be ranked up;
Wherein, as shown in fig. 7, described device also includes:
Recommendation unit 702, recommends the first application App, wherein N for coming top n the first son application
For presetting positive integer;Or, first son application of the recommendation degree score more than predetermined threshold value is recommended described first
Application App.
Alternatively, as shown in figure 8, the processing unit 604 can include that the first computing module 802, second is calculated
Module 804 and the 3rd computing module 806.
Wherein, the first computing module 802, for according to the operation behavior information and the operation behavior information
Weight, calculates the first score value of each the first son application;Second computing module 804, for belonging to the user
Property information mated with the application feature database for pre-building, calculate second score value of each the first son application;The
Three computing modules 806, for asking with second score value to first score value of each the first son application
With obtain each corresponding score value of the first son application described.
Herein it should be noted that above-mentioned first computing module 802, the second computing module 804 and the 3rd calculate mould
Block 806 corresponding to step S3022 in embodiment one to step S3026, realized with corresponding step by three modules
Example identical with application scenarios, but be not limited to one disclosure of that of above-described embodiment.It should be noted that above-mentioned
Module is may operate in the terminal 10 of the offer of embodiment one as a part for device, can pass through software reality
Existing, it is also possible to be realized by hardware.
Alternatively, first computing module 802 is used for executing following steps according to the operation behavior information and institute
The weight of operation behavior information is stated, the first score value of each the first son application is calculated:To the operation behavior information
It is normalized;By formulaFirst score value of each the first son application is calculated,
Wherein, S1 represents first score value, YiRepresent that the operation behavior information after normalization, n represent the operation row
For the number of information, XiRepresent the weight of the operation behavior information.
Alternatively, the operation behavior information includes the following combination of one or more:Last time use time, point
Hit number of times and payment times.
Alternatively, in the case where the operation behavior information includes the last time use time, first meter
Calculating module 802 includes:First sub-acquisition module, for obtaining in the advance user behavior sample set for gathering corresponding to institute
State maximum and the minima of last time use time;First sub- computing module, for passing through formula Y1=(R-Rmax)
/(Rmax-Rmin) calculate the last time use time after normalization, wherein, Y1After representing the normalization
The last time use time, R represents the last time use time, RmaxRepresent corresponding to described last
The maximum of first use time, RminRepresent the minima corresponding to the last time use time.
Alternatively, in the case where the operation behavior information includes the number of clicks, first computing module 802
Including:Second sub-acquisition module, for obtaining in the user behavior sample set of the advance collection corresponding to the click
The maximum of number of times and minima;Second sub- computing module, for passing through formula Y2=(F-Fmin)/(Fmax-Fmin)
Calculate the number of clicks after normalization, wherein, Y2Represent that the number of clicks after the normalization, F are represented
The number of clicks, FmaxRepresent the maximum corresponding to the number of clicks, FminRepresent corresponding to the click time
Several minima;
Alternatively, in the case where the operation behavior information includes the payment times, first computing module 802
Including:3rd sub-acquisition module, for obtaining in the user behavior sample set of the advance collection corresponding to the payment
The maximum of number of times and minima;3rd sub- computing module, for passing through formula Y3=(M-Mmin)/(Mmax-Mmin)
Calculate the payment times after normalization, wherein, Y3Represent that the payment times after the normalization, F are represented
The payment times, MmaxRepresent the maximum corresponding to the payment times, MminRepresent corresponding to the payment
The minima of number of times.
Alternatively, as shown in figure 9, second computing module 804 can include matched sub-block 902, assignment
Module 904 and calculating sub module 906.
Wherein, matched sub-block 902, are belonged to the user for searching in the application feature database for pre-building
The application of property information match;Assignment submodule 904, for described match with the customer attribute information should
With imparting preset fraction;Calculating sub module 906, should for according to the preset fraction, calculating each first son described
Second score value.
Herein it should be noted that above-mentioned matched sub-block 902, assignment submodule 904 and calculating sub module 906
The example that is realized to step S24, the module with corresponding step corresponding to step S20 in embodiment one and application
Scene is identical, but is not limited to one disclosure of that of above-described embodiment.It should be noted that above-mentioned module is used as device
A part may operate in the terminal 10 of the offer of embodiment one, can be realized by software, it is also possible to logical
Cross hardware realization.
Alternatively, as shown in Figure 10, described device can also include:Determining unit 1002 and assignment unit 1004.
Wherein it is determined that unit 1002, for determining that the first application App does not generate institute in the first preset time period
Stating operation information and described first applies the second son that App generates the operation information in the second preset time period to answer
With;Assignment unit 1004, for giving default score value to the described second son application;Wherein, the sequencing unit 606
For executing recommendation degree score of the following steps according to each the first son application, recommend to the described first application App
The first son application:According to score value order from big to small, first son is recommended to answer to the described first application App
With and described second son application.
Herein it should be noted that above-mentioned determining unit 1002 and assignment unit 1004 are corresponding to the step in embodiment one
, to step S404, the module is identical with example and application scenarios that corresponding step is realized, but is not limited to for rapid S402
One disclosure of that of above-described embodiment.It should be noted that above-mentioned module is may operate in as a part for device
In the terminal 10 that embodiment one is provided, can be realized by software, it is also possible to realized by hardware.
Alternatively, as shown in figure 11, described device can also include:3rd acquiring unit 1102.
Wherein, the 3rd acquiring unit 1102, for obtaining the 3rd son that need to be arranged in before each the first son application described
Application;Wherein, the recommendation unit 606 is used for executing recommendation degree of the following steps according to each the first son application
Score, recommends the first son application to the described first application App:To the described 3rd son application and described each the
With the described 3rd son application identical application during one son application is ranked up, and each first son is applied described in deleting;To
The first application App recommends the 3rd son application and each the first son application described, wherein, each first son described
Do not include in application and the described 3rd son application identical application.
Herein it should be noted that above-mentioned 3rd acquiring unit 1102 is corresponding to step S30 in embodiment one, the mould
Block is identical with example and application scenarios that corresponding step is realized, but is not limited to one disclosure of that of above-described embodiment.
It should be noted that above-mentioned module may operate in the terminal 10 that embodiment one is provided as a part for device
In, can be realized by software, it is also possible to realized by hardware.
Alternatively, as shown in figure 12, described device can also include:4th acquiring unit 1202.
Wherein, the 4th acquiring unit 1202, for obtaining the 4th son that need to be arranged in after each first son described is applied
Application;Wherein, the sequencing unit 606 is used for executing recommendation degree of the following steps according to each the first son application
Score, recommends the first son application to the described first application App:To the described 4th son application and described each the
One son application be ranked up, and delete described 4th son application in described each first son application identical application;To
The first application App recommends the 4th son application and each the first son application described, wherein, the 4th son application
In not comprising and described each first son application identical application.
Herein it should be noted that above-mentioned 4th acquiring unit 1202 is corresponding to step S40 in embodiment one, the mould
Block is identical with example and application scenarios that corresponding step is realized, but is not limited to one disclosure of that of above-described embodiment.
It should be noted that above-mentioned module may operate in the terminal 10 that embodiment one is provided as a part for device
In, can be realized by software, it is also possible to realized by hardware.
Alternatively, the customer attribute information includes the following combination of one or more:Occupation, age, sex, disappear
Take type and consumption degree.
Embodiment 3
Embodiments of the invention additionally provide a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium
Can be used for preserving the program code performed by the sort method of the application App neutron applications provided by above-described embodiment one.
Alternatively, in the present embodiment, above-mentioned storage medium is may be located in computer network Computer terminal group
In any one terminal, or in any one mobile terminal in mobile terminal group.
Alternatively, in the present embodiment, storage medium is arranged to store the program code for being used for executing following steps:
The customer attribute information using the first application App is obtained, wherein, the customer attribute information is used for the spy of instruction user
The information that levies, the first application App include multiple first son applications;By the customer attribute information with each the
The application feature database that one son application pre-builds is mated;According to the customer attribute information with application feature database
Match somebody with somebody result, each the first son application described is ranked up.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Calculate respectively described
The information similarity of the application feature database that customer attribute information is pre-build with each the first son application, described information are similar
Spend the matching result with each application feature database for the customer attribute information.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Obtain described first
The operation information of each first son application of the user of application App in the first preset time period, wherein, the operation
Information includes operation behavior information of the user of the first application App in each the first son application.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:According to the operation
Behavioural information, the weight of the operation behavior information and the customer attribute information, be calculated respectively each first
The recommendation degree score of son application.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:According to described each
The recommendation degree score of the first son application, is ranked up to each the first son application described;According to each first son described
The recommendation degree score of application, after being ranked up to each the first son application described, methods described also includes:To come
The first application App is recommended in top n the first son application, and wherein N is default positive integer;Or, push away described
Degree of recommending score recommends the first application App more than the first son application of predetermined threshold value.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:According to the operation
Behavioural information and the weight of the operation behavior information, calculate the first score value of each the first son application;By institute
State customer attribute information and mated with the application feature database for pre-building, calculate the second of each the first son application
Score value;First score value of each the first son application is sued for peace with second score value, is obtained described each
Individual first son applies the score value of corresponding recommendation degree score.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:The operation is gone
It is normalized for information;By formulaCalculate described the first of each the first son application
Score value, wherein, S1 represents first score value, YiRepresent that the operation behavior information after normalization, n represent described
The number of operation behavior information, XiRepresent the weight of the operation behavior information.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:In the operation row
In the case of including the last time use time for information, described place is normalized to the operation behavior information
Reason, including:Obtain the maximum corresponding to the last time use time in the user behavior sample set of collection in advance
And minima;By formula Y1=(R-Rmax)/(Rmax-Rmin) calculate the last time use after normalization
Time, wherein, Y1Represent that the last time use time after the normalization, R represent that the last time makes
With time, RmaxRepresent the maximum corresponding to the last time use time, RminRepresent corresponding to described last
The minima of first use time;In the case where the operation behavior information includes the number of clicks, described to institute
State operation behavior information to be normalized, including:Obtain corresponding in the user behavior sample set of the advance collection
Maximum and minima in the number of clicks;By formula Y2=(F-Fmin)/(Fmax-Fmin) calculate normalizing
The number of clicks after change, wherein, Y2Represent that the number of clicks after the normalization, F represent the click
Number of times, FmaxRepresent the maximum corresponding to the number of clicks, FminRepresent the minimum corresponding to the number of clicks
Value;In the case where the operation behavior information includes the payment times, described the operation behavior information is carried out
Normalized, including:Obtain in the user behavior sample set of the advance collection and correspond to the payment times most
Big value and minima;By formula Y3=(M-Mmin)/(Mmax-Mmin) calculate the payment time after normalization
Number, wherein, Y3Represent that the payment times after the normalization, F represent the payment times, MmaxIt is right to represent
The maximum of payment times described in Ying Yu, MminRepresent the minima corresponding to the payment times.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Built described in advance
Search, in vertical application feature database, the application matched with the customer attribute information;Believe with the user property to described
The application of manner of breathing coupling gives preset fraction;According to the preset fraction, the described of each the first son application is calculated
Second score value.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Determine described first
Application App does not generate the operation information in the first preset time period and the first application App is default second
The second son application of the operation information is generated in time period;Default score value is given to the described second son application;Wherein,
The recommendation degree score of each the first son application described in the basis, should to the described first application App recommendations first son
With, including:According to score value order from big to small, to the described first application App recommend first son application and
The second son application.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Obtain and need to be arranged in
The 3rd son application before each the first son application described;Wherein, the recommendation of each the first son application described in the basis
Degree score, recommends the first son application to the described first application App, including:To the described 3rd son application and institute
State each the first son application to be ranked up, and delete in each the first son application described with the described 3rd son application identical
Application;Recommend the 3rd son application and each the first son application described to the described first application App, wherein, described each
Do not include in individual first son application and the described 3rd son application identical application.
Alternatively, storage medium is also configured to store the program code for being used for executing following steps:Obtain and need to be arranged in
The 4th son application after each the first son application described;Wherein, the recommendation of each the first son application described in the basis
Degree score, recommends the first son application to the described first application App, including:To the described 4th son application and institute
State each first son application be ranked up, and delete described 4th son application in described each first son application identical
Application;Recommend the 4th son application and each the first son application described, wherein, described the to the described first application App
Do not include in four son applications and each the first son application identical application described.
Alternatively, in the present embodiment, above-mentioned storage medium can be included but is not limited to:USB flash disk, read only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), portable hard drive, magnetic
Dish or CD etc. are various can be with the medium of store program codes.
Alternatively, the specific example in the present embodiment may be referred to the example described in above-described embodiment 1, this enforcement
Example will not be described here.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
In the above embodiment of the present invention, the description of each embodiment is all emphasized particularly on different fields, do not had in certain embodiment
The part of detailed description, may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents, other can be passed through
Mode realize.Wherein, device embodiment described above is only the schematically division of for example described unit,
It is only a kind of division of logic function, when actually realizing, can has other dividing mode, for example multiple units or component
Can in conjunction with or be desirably integrated into another system, or some features can be ignored, or not execute.Another, institute
The coupling each other for showing or discussing or direct-coupling or communication connection can be by some interfaces, unit or mould
The INDIRECT COUPLING of block or communication connection, can be electrical or other forms.
The unit that illustrates as separating component can be or may not be physically separate, aobvious as unit
The part for showing can be or may not be physical location, you can be located at a place, or can also be distributed to
On multiple NEs.Some or all of unit therein can be selected according to the actual needs to realize the present embodiment
The purpose of scheme.
In addition, each functional unit in each embodiment of the invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.Above-mentioned integrated
Unit both can be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If the integrated unit realized using in the form of SFU software functional unit and as independent production marketing or use when,
Can be stored in a computer read/write memory medium.It is based on such understanding, technical scheme essence
On all or part of part that in other words prior art is contributed or the technical scheme can be with software product
Form is embodied, and the computer software product is stored in a storage medium, is used so that one including some instructions
Platform computer equipment (can be personal computer, server or network equipment etc.) executes each embodiment institute of the invention
State all or part of step of method.And aforesaid storage medium includes:USB flash disk, read only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), portable hard drive, magnetic disc or CD
Etc. various can be with the medium of store program codes.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improve and moisten
Decorations also should be regarded as protection scope of the present invention.