CN108564220A - User gender prediction method, apparatus, storage medium and computer equipment - Google Patents
User gender prediction method, apparatus, storage medium and computer equipment Download PDFInfo
- Publication number
- CN108564220A CN108564220A CN201810354699.XA CN201810354699A CN108564220A CN 108564220 A CN108564220 A CN 108564220A CN 201810354699 A CN201810354699 A CN 201810354699A CN 108564220 A CN108564220 A CN 108564220A
- Authority
- CN
- China
- Prior art keywords
- application
- gender
- user
- target user
- list
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3438—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment monitoring of user actions
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Hardware Design (AREA)
- Development Economics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of user gender prediction method, and this method includes：Obtain the application using pond；Obtain the list of application of target user；According to the application and the list of application, first eigenvector is generated；The first eigenvector is inputted preset gender prediction's model, obtains the predicted value of target user's gender；Gender prediction's model is used to characterize the incidence relation of first eigenvector and user's gender；According to the predicted value, the gender of the target user is predicted.The gender prediction of target user is realized using gender prediction's model of advance training completion, in user in the case where application platform does not have any gender relevant information to reveal, also the gender of target user is predicted, and then is oriented for the gender according to user in application scenarios and provides strong support using the application dispensing installation conversion ratio for launching, improving application platform.
Description
Technical field
The present invention relates to field of computer technology, specifically, the present invention relates to a kind of user gender prediction method, dresses
It sets, computer readable storage medium and computer equipment.
Background technology
As the fast development of Internet technology and the quick of intelligent terminal are popularized, various application software layers go out not
Thoroughly.In order to concentrate the application download installation requirements for showing application software and meeting user to user, such as application shop or using city
Etc. can to user recommend application software and provide application software download channel application platform come into being.In application platform
In daily operation, there are many scenes to need to be oriented using dispensing according to the gender of user, such as：It is thrown surely to female user
" U.S. face camera-type application ", throws " military information class application " etc. surely to male user.
However, since application platform does not force usually to log in, most of user just directly accesses application without logging in
Platform, causes application platform that can not know the gender of these users, then cannot achieve the property according to user in application scenarios
It is not oriented using launching, the application dispensing installation conversion ratio of application platform is relatively low.
Invention content
It is that can at least solve above-mentioned one of technological deficiency, the present invention provides the user gender predictions of following technical scheme
Method, using recommending method and corresponding device, computer readable storage medium and computer equipment.
The embodiment of the present invention provides a kind of user gender prediction method according on one side, including：
Obtain the application using pond；Obtain the list of application of target user；
According to the application and the list of application, first eigenvector is generated；
The first eigenvector is inputted preset gender prediction's model, obtains the prediction of target user's gender
Value；Gender prediction's model is used to characterize the incidence relation of first eigenvector and target user's gender；
According to the predicted value, the gender of the target user is predicted.
Preferably, the list of application for obtaining target user, including：Obtain the application that target user downloads；According to mesh
Mark the list of application that the application that user downloads generates the target user；And/or
Obtain the mounted application of target user；Answering for the target user is generated according to the mounted application of target user
Use list.
Preferably, gender prediction's model is generated by following steps：
Obtain the application using pond；Obtain the list of application of historical user；
Obtain the gender information of historical user；
According to the list of application of the application and the historical user, second feature vector is generated；
According to the corresponding gender information of the second feature vector sum, training sample is generated；
According to the training sample, gender prediction's model is generated by regression algorithm training.
Further, the list of application for obtaining historical user, including：
Obtain the application that historical user downloads；The application row of the historical user are generated according to the application that historical user downloads
Table；And/or
Obtain the mounted application of historical user；Answering for the historical user is generated according to the mounted application of historical user
Use list.
Preferably, the regression algorithm is logistic regression algorithm.
Preferably, the gender information for obtaining historical user, including：
The log-on message for obtaining historical user, gender information is extracted from the log-on message；Or
The request for obtaining gender information is sent to historical user, is carried from the feedback information that historical user responds the request
Take gender information.
Preferably, described that first eigenvector is generated according to the application and the list of application, including：
According to the application, generate comprising the first set that element is the application；
According to the list of application, it is the second set applied in the list of application to generate comprising element；
According to the first set and the second set, first eigenvector is generated using one-hot coding.
Preferably, described that the gender of the target user is predicted according to the predicted value, including：
Judge whether the predicted value is more than preset threshold value；
If so, judging the gender of the target user for male；
If it is not, judging the gender of the target user for women.
The embodiment of the present invention additionally provides a kind of application recommendation method according to one side, including：
According to above-mentioned user's gender prediction's method, the gender of target user is predicted；
It is obtained from application pond preset with the matched application to be recommended of the gender；
Recommend the application to be recommended to the target user.
In addition, the embodiment of the present invention provides a kind of user gender prediction device according on the other hand, including：
Using acquisition module, for obtaining the application using pond；Obtain the list of application of target user；
Feature vector generation module, for according to the application and the list of application, generating first eigenvector；
Predictor calculation module obtains described for the first eigenvector being inputted preset gender prediction's model
The predicted value of target user's gender；
Gender prediction's module, for according to the predicted value, predicting the gender of the target user.
The embodiment of the present invention is a kind of using recommendation apparatus according on the other hand, additionally providing, including：
Gender prediction's module, for according to above-mentioned user gender prediction method, predicting the gender of target user；
It is preset with the matched application to be recommended of the gender for being obtained from application pond using acquisition module；
Using recommending module, for recommending the application to be recommended to the target user.
The embodiment of the present invention provides a kind of computer readable storage medium, the computer according to another aspect
Computer program is stored on readable storage medium storing program for executing, the computer program realizes above-mentioned user's gender when being executed by processor
Prediction technique.
The embodiment of the present invention additionally provides a kind of computer readable storage medium, the calculating according to another aspect
Computer program is stored on machine readable storage medium storing program for executing, the computer program realizes that above-mentioned application pushes away when being executed by processor
Recommend method.
The embodiment of the present invention provides a kind of computer equipment according to another aspect, and the computer includes one
Or multiple processors；Memory；One or more computer programs, wherein one or more of computer programs are stored in
It in the memory and is configured as being executed by one or more of processors, one or more of computer program configurations
For：Execute above-mentioned user's gender prediction's method.
The embodiment of the present invention additionally provides a kind of computer equipment, the computer equipment packet according to another aspect
Include one or more processors；Memory；One or more computer programs, wherein one or more of computer program quilts
It is stored in the memory and is configured as being executed by one or more of processors, one or more of computer journeys
Sequence is configured to：Execute above-mentioned application recommendation method.
Compared with prior art, the present invention having the advantages that：
User gender prediction method, apparatus, storage medium and computer equipment provided by the invention, based on target user's
Application message is realized the gender prediction of target user using gender prediction's model of advance training completion, is being applied in user
In the case that platform does not have any gender relevant information to reveal, the gender of target user is also predicted, and then in application scenarios
The middle gender according to user is oriented provides strong branch using the application dispensing installation conversion ratio for launching, improving application platform
It holds.
In addition, method, apparatus, storage medium and computer equipment are recommended in application provided by the invention, carry through the invention
Gender prediction's method of confession realizes the gender prediction of target user, so realize in application scenarios according to the gender of user into
Row orientation application is launched, and installation conversion ratio is launched in the application for being remarkably improved application platform.
The additional aspect of the present invention and advantage will be set forth in part in the description, these will become from the following description
Obviously, or practice through the invention is recognized.
Description of the drawings
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, wherein：
Fig. 1 is the method flow diagram of user gender prediction method provided in an embodiment of the present invention；
Fig. 2 is the method flow diagram provided in an embodiment of the present invention for generating gender prediction model；
Fig. 3 is the method flow diagram that method is recommended in application provided in an embodiment of the present invention；
Fig. 4 is the structural schematic diagram of user gender prediction device provided in an embodiment of the present invention；
Fig. 5 is the structural schematic diagram provided in an embodiment of the present invention using recommendation apparatus；
Fig. 6 is the structural schematic diagram of computer equipment provided in an embodiment of the present invention.
Specific implementation mode
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and is only used for explaining the present invention, and is not construed as limiting the claims.
Embodiment one
An embodiment of the present invention provides a kind of user gender prediction methods, as shown in Figure 1, this method includes：
Step S101：Obtain the application using pond；Obtain the list of application of target user.
For the present embodiment, it can recommend application software to user in such as application shop or application market etc. and provide using soft
Part is downloaded in the application platform of channel, includes various application software, the class software of such as doing shopping, audio broadcast message class software,
Game class software etc..
For the present embodiment, the application using pond is presently described application platform, in application shop, application market
All application software.
Wherein, the list of application for obtaining target user, including：Obtain the application that target user downloads；According to target
The application that user downloads generates the list of application of the target user；And/or obtain the mounted application of target user；According to
The mounted application of target user generates the list of application of the target user.
For the present embodiment, obtaining target user's record has the user behaviors log of its application and mounted application downloaded,
The application and/or mounted application that target user downloads are obtained from the user behaviors log.The list of application can be mesh
The download list of application for marking user, can also be the installation list of application of target user, can also be comprising under target user
It carries application and the list of application of application has been installed, wherein the download list of application is that target user is preset in application platform
The list of the application software downloaded downloaded in time or accumulative, the installation list of application are that target user is currently mounted
The list of application software.In practical application scene, downloading the application that user has downloaded in list of application may be unloaded by user
It carries, therefore the installation list of application of user and not comprising such application being unloaded.When the list of application is to download application row
When table, the application that can be downloaded by obtaining target user generates answering for the target user according to the application that target user downloads
Use list；It, can be by obtaining the mounted application of target user, according to mesh when the list of application is to have installed list of application
The mark mounted application of user generates the list of application of the target user；When the list of application includes that target user's download is answered
With with application has been installed list of application when, can by obtain target user download application and mounted application, to target
The list of application of the target user is generated after the application of user's download and mounted application progress duplicate removal.
Step S102：According to the application and the list of application, first eigenvector is generated.
For the present embodiment, according to the list of application of the application and target user in application pond, generate for pre- as gender
The first eigenvector of mode input characteristic variable is surveyed, the first eigenvector can embody the application feature of target user.Its
In, the first eigenvector is a n-dimensional vector, and the n is identical as the number of the application using pond.
Step S103：The first eigenvector is inputted preset gender prediction's model, obtains target user's property
Other predicted value；Gender prediction's model is used to characterize the incidence relation of first eigenvector and target user's gender.
For the present embodiment, gender prediction's model is used to characterize the first eigenvector and target user's gender
Incidence relation inputs trained gender prediction's model in advance using as the first eigenvector of input feature vector variable,
It can obtain the predicted value of the gender according to the calculated target user of the first eigenvector.
Step S104：According to the predicted value, the gender of the target user is predicted.
For the present embodiment, according to the predicted value, predict that the gender of the target user is male or women.Specifically
Then predict that the gender of the target user is when the predicted value meets the condition for judging the target user for male in ground
Male；When the predicted value meets the condition for judging the target user for women, then the gender of the target user is predicted
For women.
In user gender prediction method provided in an embodiment of the present invention, the application message based on target user, using pre-
It first trains the gender prediction's model completed to realize the gender prediction of target user, there is no any gender in application platform in user
In the case that relevant information is revealed, also predict the gender of target user, so in application scenarios according to the gender of user
It is oriented and provides strong support using the application dispensing installation conversion ratio for launching, improving application platform.
Embodiment two
The alternatively possible realization method of the embodiment of the present invention further includes implementing on the basis of shown in embodiment one
Step shown in example two, wherein
As shown in Fig. 2, gender prediction's model described in the step S103 is generated by following steps：
Step S201：Obtain the application using pond；Obtain the list of application of historical user.
For the present embodiment, it can recommend application software to user in such as application shop or application market etc. and provide using soft
Part is downloaded in the application platform of channel, includes various application software, the class software of such as doing shopping, audio broadcast message class software,
Game class software etc..
For the present embodiment, the application using pond is presently described application platform, in application shop, application market
All application software.
For the present embodiment, the historical user indicate to have in application platform in preset time or in application platform once
Execute the validated user of application download and/or application installation behavior.Wherein, the preset time can be one day, one week, 30 days
Equal times, the present embodiment do not limit this.
Wherein, the list of application for obtaining historical user, including：Obtain the application that historical user downloads；According to history
The application that user downloads generates the list of application of the historical user；And/or obtain the mounted application of historical user；According to
The mounted application of historical user generates the list of application of the historical user.
For the present embodiment, obtaining historical user's record has the user behaviors log of its application and mounted application downloaded,
The application and/or mounted application that historical user downloads are obtained from the user behaviors log.The list of application can be gone through
The download list of application of history user can also be the installation list of application of historical user, can also be comprising under historical user
It carries application and the list of application of application has been installed, wherein the download list of application is that historical user is preset in application platform
The list of the application software downloaded downloaded in time or accumulative, the installation list of application are that historical user is currently mounted
The list of application software.In practical application scene, downloading the application that user has downloaded in list of application may be unloaded by user
It carries, therefore the installation list of application of user and not comprising such application being unloaded.When the list of application is to download application row
When table, the application that can be downloaded by obtaining historical user generates answering for the historical user according to the application that historical user downloads
Use list；It, can be by obtaining the mounted application of historical user, according to going through when the list of application is to have installed list of application
The mounted application of history user generates the list of application of the historical user；When the list of application includes that historical user's download is answered
With with application has been installed list of application when, can by obtain historical user download application and mounted application, to history
The list of application of the historical user is generated after the application of user's download and mounted application progress duplicate removal.
Step S202：Obtain the gender information of historical user.
For the present embodiment, the application platform knows the gender information of the historical user in advance, has from record described
The gender information of historical user is obtained in the data information of gender information.
Specifically, the gender information of historical user can be obtained by following two modes.
Mode one：The log-on message for obtaining historical user, gender information is extracted from the log-on message.For we
Formula, the historical user are the user for having in application platform registration login behavior, and the historical user registers in application platform
When typing have its gender information.Log-on message by obtaining historical user is gone through, it can be achieved that being extracted from the log-on message
The gender information of history user.
Mode two：The request for obtaining gender information is sent to historical user, and the feedback of the request is responded from historical user
Gender information is extracted in information.It, can be being sent to user in the form of transmission questionnaire etc. and obtain gender information for the method
Request extracts gender for there is the user for responding the request, sending feedback information to application platform from its feedback information
Information.
Need it is clear that, the mode of gender information for obtaining historical user is not limited by the present embodiment,
It can also realize that the gender information of historical user obtains by other means.
Step S203：According to the list of application of the application and the historical user, second feature vector is generated.
For the present embodiment, according to the list of application of the application and historical user in application pond, generate for pre- as gender
The second feature vector of mode input characteristic variable is surveyed, the second feature vector can embody the application feature of historical user.Its
In, the second feature vector is a n-dimensional vector, and the n is identical as the number of the application using pond.
Step S204：According to the corresponding gender information of the second feature vector sum, training sample is generated.
For the present embodiment, after obtaining the gender information, the gender information is also needed to be converted into corresponding history use
The predicted value of family gender, the predicted value are 1 dimensional vector.For example, pre-setting 1 expression male, 0 indicates women, due to obtaining
It is the fixed gender information of user, then when the gender of historical user is male, the predicted value is 1, works as historical user
Gender be women when, the predicted value be 0.
It is vectorial according to the second feature for the input feature vector variable as gender prediction's model for the present embodiment, with
And the predicted value of the corresponding gender information of historical user belonging to the second feature vector, it generates for training gender prediction's model
Training sample.
For the present embodiment, the historical user include have in application platform in preset time or in application platform once
Execute multiple validated users of application download and/or application installation behavior.Therefore according to the present embodiment above-mentioned steps also produce pair
Answer multiple training samples of multiple validated users.
Step S205：According to the training sample, gender prediction's model is generated by regression algorithm training.
For the present embodiment, as a preferred embodiment, the regression algorithm is logistic regression algorithm.The logic
The normalized form of regression algorithm is：
Wherein, x indicates that second feature vector, the second feature vector are the vectors of n dimensions；Y is indicated for predicting
The predicted value of user's gender is 1 dimensional vector, is preset when the gender of user is male, and the predicted value is 1,1 expression man
Property, when the gender of user is women, the predicted value is 0,0 expression women；W indicates different characteristic pair in logistic regression algorithm
The weight answered is a n-dimensional vector；B indicates the constant term in logistic regression algorithm.
For the present embodiment, by by multiple training samples for including the corresponding predicted value of the second feature vector sum
Originally it is input in the normalized form of above-mentioned logistic regression algorithm, and gender prediction's model loss function is solved by gradient descent method
Homographic solution w, b when los (w, b) is minimum obtain parameter w, the b of gender prediction's model, and then training generates gender prediction's mould
Type.
Need it is clear that, it is above-mentioned that generate gender prediction's model using the training of logistic regression algorithm be only one kind
The method for obtaining gender prediction model can also use the training of other algorithms to generate in addition to using above-mentioned logistic regression algorithm
Gender prediction's model.
In user gender prediction method provided in an embodiment of the present invention, the application message based on historical user and its gender
Information, training generate can be used for realize target user gender prediction gender prediction's model, in application scenarios according to
The gender at family is oriented provides strong support using the application dispensing installation conversion ratio for launching, improving application platform.
Embodiment three
The alternatively possible realization method of the embodiment of the present invention further includes implementing on the basis of shown in embodiment one
Step shown in example three, wherein
The step S102, including：
According to the application, generate comprising the first set that element is the application；According to the list of application, packet is generated
It is the second set applied in the list of application containing element；According to the first set and the second set, using only heat
Coding generates first eigenvector.
For the present embodiment, after obtaining the application using pond in the step S101, according to the application, the is generated
One set generates the set of applications using pond；Wherein, the element in the first set is corresponded to apply and respectively be applied in pond, institute
State element number in first set with it is described identical using the application number in pond.
For the present embodiment, after the list of application that target user is obtained in the step S101, according to the application
List generates second set, that is, generates the set of applications of target user.When the list of application is multiple lists, according to more
Application in a list of application, by using generating the second set after duplicate removal.
It is described according to the first set and the second set for the present embodiment, generate first using one-hot coding
Feature vector, specially：First, after generating the first set, it is pressed in set to the application in the first set
Put in order and be numbered, if the first set has a n element, the serial number applied in the first set 1 to n it
Between；Then, one-hot one-hot codings are carried out to the second set that application is downloaded and/or installed comprising target user, initially
Change a n-dimensional vector, will be using the serial number position value of the corresponding first set in second set described in the n-dimensional vector
1, remaining value is 0, finally obtains the first eigenvector generated using one-hot coding.
For example, being appList={ A, B, C, D } according to the second set that the list of application of target user generates；Wherein, it answers
It is corresponding with the application of serial number in the first set 1,3,4,6 respectively with A, B, C, D, then the n-dimensional vector serial number 1,
3,4,6 values are 1, remaining value is 0, then according to the first set and the second set, using one-hot coding
The first eigenvector of generation is X=(1,0,1,1,0,1 ..., 0).
For the present embodiment, according to the list of application of the application and target user in application pond, generate for pre- as gender
The first eigenvector of mode input characteristic variable is surveyed, the first eigenvector can embody the application feature of target user.Its
In, the first eigenvector is a n-dimensional vector, and the n is identical as the number of the application using pond.
Example IV
The alternatively possible realization method of the embodiment of the present invention further includes implementing on the basis of shown in embodiment one
Step shown in example four, wherein
The step S104, including：
Judge whether the predicted value is more than preset threshold value；If so, judging the gender of the target user for male；If
It is no, judge the gender of the target user for women.
For the present embodiment, by pre-setting threshold value, by the predicted value being calculated in the step S103 with
The preset threshold value is compared, and then judges the gender of target user.
For example, pre-setting 1 expression male, 0 indicates that women, the preset threshold value are 0.5, when the predicted value is more than institute
When stating preset threshold value 0.5, judge the gender of the target user for male；When the predicted value is less than or equal to the preset threshold value
When 0.5, judge the gender of the target user for women.
, it can be achieved that predicted value obtained by calculation is sentenced in the user gender prediction method provided in an embodiment of the present invention
The gender of disconnected target user, to be oriented using dispensing according to the gender of user in application scenarios, improving application platform
Strong support is provided using installation conversion ratio is launched.
In addition, an embodiment of the present invention provides a kind of applications to recommend method, as shown in figure 3, this method includes：
Step S301：According to user's gender prediction's method described in one embodiment of any of the above, the gender of target user is predicted.
For the present embodiment, can be realized by the embodiment of the method for above-mentioned any one of user gender prediction method
The gender prediction of target user, specific implementation is referring to the explanation of above method embodiment, and details are not described herein.
Step S302：It is obtained from application pond preset with the matched application to be recommended of the gender.
For the present embodiment, preset in the application platform with the matched application to be recommended of user's gender, for example, preset
It is described with the matched application to be recommended of male include such as military affairs information class, crime reasoning class application, preset described and female
Property matched application to be recommended include just like applications such as U.S. face camera-type, fashion bull's classes.It is measured in advance in the step S301
After the gender of target user, corresponding one or more application to be recommended is obtained according to the gender.
Step S303：Recommend the application to be recommended to the target user.
For the present embodiment, recommend acquired one or more application to be recommended to the target user.
Application provided by the invention recommends method, and the gender prediction's method provided through the invention realizes target user's
Gender prediction, and then realize and be oriented according to the gender of user using launching in application scenarios, it is remarkably improved using flat
Installation conversion ratio is launched in the application of platform.
In addition, an embodiment of the present invention provides a kind of user gender prediction devices, as shown in figure 4, described device includes：
Using acquisition module 401, for obtaining the application using pond；Obtain the list of application of target user.
Feature vector generation module 402, for according to the application and the list of application, generating first eigenvector.
Predictor calculation module 403 obtains institute for the first eigenvector being inputted preset gender prediction's model
State the predicted value of target user's gender.
Gender prediction's module 404, for according to the predicted value, predicting the gender of the target user.
User gender prediction device provided by the invention, it can be achieved that：Application message based on target user, using advance instruction
Practice gender prediction's model of completion to realize the gender prediction of target user, there is no any gender related in application platform in user
In the case that information is revealed, the gender of target user is also predicted, and then to be carried out according to the gender of user in application scenarios
Orientation application is launched, the application of raising application platform launches installation conversion ratio and provides strong support.
The embodiment of the method for above-mentioned offer, specific work(may be implemented in user gender prediction device provided in an embodiment of the present invention
It can realize the explanation referred in embodiment of the method, details are not described herein.
In addition, the embodiment of the present invention additionally provide it is a kind of apply recommendation apparatus, as shown in figure 5, described device includes：
Gender prediction's module 501, for according to user gender prediction method described in any of the above embodiment, prediction target to be used
The gender at family；
It is preset with the matched application to be recommended of the gender for being obtained from application pond using acquisition module 502；
Using recommending module 503, for recommending the application to be recommended to the target user.
Application recommendation apparatus provided by the invention, it can be achieved that：Gender prediction's method for there is provided through the invention realizes mesh
The gender prediction of user is marked, and then realizes and is oriented according to the gender of user using launching in application scenarios, can significantly be carried
Installation conversion ratio is launched in the application of high application platform.
The embodiment of the method provided in an embodiment of the present invention that above-mentioned offer may be implemented using recommendation apparatus, concrete function are real
Referring now to the explanation in embodiment of the method, details are not described herein.
In addition, an embodiment of the present invention provides a kind of computer readable storage medium, deposited on computer readable storage medium
Computer program is contained, user's gender described in above example one to four is realized when the computer program is executed by processor
Prediction technique and above application recommend method.Wherein, the computer readable storage medium is including but not limited to any kind of
Disk (including floppy disk, hard disk, CD, CD-ROM and magneto-optic disk), ROM (Read-Only Memory, read-only memory), RAM
(Random AcceSS Memory, immediately memory), EPROM (EraSable Programmable Read-Only
Memory, Erarable Programmable Read only Memory), EEPROM (Electrically EraSable Programmable
Read-Only Memory, Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or light card.It is, storage
Equipment includes by any medium of equipment (for example, computer, mobile phone) storage or transmission information in the form of it can read, Ke Yishi
Read-only memory, disk or CD etc..
Computer readable storage medium provided by the invention, it can be achieved that：Application message based on target user, using advance
It trains the gender prediction's model completed to realize the gender prediction of target user, there is no any gender phase in application platform in user
Close in the case that information reveals, also predict the gender of target user, so in application scenarios according to the gender of user into
Row orientation application is launched, the application of raising application platform launches installation conversion ratio and provides strong support.It can also be achieved：Pass through this
Gender prediction's method of offer is invented to realize the gender prediction of target user, and then is realized in application scenarios according to user's
Gender is oriented the application dispensing installation conversion ratio that application platform is remarkably improved using launching.
The embodiment of the method for above-mentioned offer may be implemented in computer readable storage medium provided in an embodiment of the present invention, specifically
Function realizes the explanation referred in embodiment of the method, and details are not described herein.In addition, the embodiment of the present invention additionally provides a kind of meter
Machine equipment is calculated, as shown in Figure 6.Computer equipment described in the present embodiment can be that server, personal computer and network are set
The equipment such as standby.The computer equipment includes the devices such as processor 602, memory 603, input unit 604 and display unit 605
Part.It, can be with it will be understood by those skilled in the art that the device structure device shown in Fig. 6 does not constitute the restriction to all devices
Including than illustrating more or fewer components, or the certain components of combination.Memory 603 can be used for storing computer program 601
And each function module, processor 602 runs the computer program 601 for being stored in memory 603, to execute the various of equipment
Application of function and data processing.Memory can be built-in storage or external memory, or including built-in storage and external storage
Both devices.Built-in storage may include read-only memory (ROM), programming ROM (PROM), electrically programmable ROM (EPROM), electricity
Erasable programmable ROM (EEPROM), flash memory or random access memory.External memory may include hard disk, floppy disk,
ZIP disks, USB flash disk, tape etc..Memory disclosed in this invention includes but not limited to the memory of these types.Institute of the invention is public
The memory opened is only used as example rather than as restriction.
Input unit 604 is used to receive the input of signal, and receives keyword input by user.Input unit 604 can
Including touch panel and other input equipments.Touch panel collects user on it or neighbouring touch operation (for example is used
Family uses the operations of any suitable object or attachment on touch panel or near touch panel such as finger, stylus), and root
According to the corresponding attachment device of preset driven by program；Other input equipments can include but is not limited to physical keyboard, function
It is one or more in key (such as broadcasting control button, switch key etc.), trace ball, mouse, operating lever etc..Display unit
605 can be used for showing information input by user or be supplied to the information of user and the various menus of computer equipment.Display is single
The forms such as liquid crystal display, Organic Light Emitting Diode can be used in member 605.Processor 602 is the control centre of computer equipment, profit
With the various pieces of various interfaces and the entire computer of connection, by running or executing the software being stored in memory 602
Program and/or module, and the data being stored in memory are called, perform various functions and handle data.
As one embodiment, the computer equipment includes：One or more processors 602, memory 603, one
Or multiple computer programs 601, wherein one or more of computer programs 601 be stored in memory 603 and by with
It is set to and is executed by one or more of processors 602, one or more of computer programs 601 are configured to carry out above
User's gender prediction's method described in embodiment one to four.
As another embodiment, the computer equipment includes：One or more processors 602, memory 603, one
A or multiple computer programs 601, wherein one or more of computer programs 601 are stored in memory 603 and quilt
Be configured to be executed by one or more of processors 602, one or more of computer programs 601 be configured to carry out with
Recommendation method is applied described in upper.
Computer equipment provided by the invention, it can be achieved that：Application message based on target user is completed using advance training
Gender prediction's model realize the gender prediction of target user, there is no any gender relevant information saturating in application platform in user
In the case of dew, the gender of target user is also predicted, and then answer to be oriented according to the gender of user in application scenarios
Installation conversion ratio is launched with the application for launching, improving application platform, and strong support is provided.It can also be achieved：It provides through the invention
Gender prediction's method realize the gender prediction of target user, and then realize and carried out according to the gender of user in application scenarios
Orientation application is launched, and installation conversion ratio is launched in the application for being remarkably improved application platform.
The embodiment of the method for above-mentioned offer may be implemented in computer equipment provided in an embodiment of the present invention, and concrete function is realized
The explanation in embodiment of the method is referred to, details are not described herein.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, it can also
That each unit physically exists alone, can also two or more units be integrated in a module.Above-mentioned integrated mould
The form that hardware had both may be used in block is realized, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and when sold or used as an independent product, can also be stored in a computer
In read/write memory medium.
The above is only some embodiments of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (15)
1. a kind of user gender prediction method, which is characterized in that include the following steps：
Obtain the application using pond；Obtain the list of application of target user；
According to the application and the list of application, first eigenvector is generated；
The first eigenvector is inputted preset gender prediction's model, obtains the predicted value of target user's gender；Property
Other prediction model is used to characterize the incidence relation of first eigenvector and target user's gender；
According to the predicted value, the gender of the target user is predicted.
2. user gender prediction method according to claim 1, which is characterized in that the application row for obtaining target user
Table, including：
Obtain the application that target user downloads；The list of application of the target user is generated according to the application that target user downloads；
And/or
Obtain the mounted application of target user；The application row of the target user are generated according to the mounted application of target user
Table.
3. user gender prediction method according to claim 1, which is characterized in that gender prediction's model passes through following
Step generates：
Obtain the application using pond；Obtain the list of application of historical user；
Obtain the gender information of historical user；
According to the list of application of the application and the historical user, second feature vector is generated；
According to the corresponding gender information of the second feature vector sum, training sample is generated；
According to the training sample, gender prediction's model is generated by regression algorithm training.
4. user gender prediction method according to claim 3, which is characterized in that the application row for obtaining historical user
Table, including：
Obtain the application that historical user downloads；The list of application of the historical user is generated according to the application that historical user downloads；
And/or
Obtain the mounted application of historical user；The application row of the historical user are generated according to the mounted application of historical user
Table.
5. user gender prediction method according to claim 3, which is characterized in that the regression algorithm is calculated for logistic regression
Method.
6. user gender prediction method according to claim 3, which is characterized in that the gender letter for obtaining historical user
Breath, including：
The log-on message for obtaining historical user, gender information is extracted from the log-on message；Or
The request for obtaining gender information, the extractability from the feedback information that historical user responds the request are sent to historical user
Other information.
7. user gender prediction method according to claim 1, which is characterized in that described to be answered with described according to the application
With list, first eigenvector is generated, including：
According to the application, generate comprising the first set that element is the application；
According to the list of application, it is the second set applied in the list of application to generate comprising element；
According to the first set and the second set, first eigenvector is generated using one-hot coding.
8. user gender prediction method according to claim 1, which is characterized in that described according to the predicted value, prediction
The gender of the target user, including：
Judge whether the predicted value is more than preset threshold value；
If so, judging the gender of the target user for male；
If it is not, judging the gender of the target user for women.
9. method is recommended in a kind of application, which is characterized in that including：
According to user gender prediction method described in claim 1 to 8 any one, the gender of target user is predicted；
It is obtained from application pond preset with the matched application to be recommended of the gender；
Recommend the application to be recommended to the target user.
10. a kind of user gender prediction device, which is characterized in that including：
Using acquisition module, for obtaining the application using pond；Obtain the list of application of target user；
Feature vector generation module, for according to the application and the list of application, generating first eigenvector；
Predictor calculation module obtains the target for the first eigenvector being inputted preset gender prediction's model
The predicted value of user's gender；
Gender prediction's module, for according to the predicted value, predicting the gender of the target user.
11. a kind of applying recommendation apparatus, which is characterized in that including：
Gender prediction's module, for according to user gender prediction method described in claim 1 to 8 any one, prediction target to be used
The gender at family；
It is preset with the matched application to be recommended of the gender for being obtained from application pond using acquisition module；
Using recommending module, for recommending the application to be recommended to the target user.
12. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program realizes claim 1 to 8 any one of them user gender prediction side when the computer program is executed by processor
Method.
13. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program is realized when the computer program is executed by processor and applies recommendation method described in claim 9.
14. a kind of computer equipment, which is characterized in that it includes：
One or more processors；
Memory；
One or more computer programs, wherein one or more of computer programs are stored in the memory and quilt
It is configured to be executed by one or more of processors, one or more of computer programs are configured to：It executes according to power
Profit requires 1 to 8 any one of them user's gender prediction's method.
15. a kind of computer equipment, which is characterized in that it includes：
One or more processors；
Memory；
One or more computer programs, wherein one or more of computer programs are stored in the memory and quilt
It is configured to be executed by one or more of processors, one or more of computer programs are configured to：It executes according to power
Profit requires to apply recommendation method described in 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810354699.XA CN108564220A (en) | 2018-04-19 | 2018-04-19 | User gender prediction method, apparatus, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810354699.XA CN108564220A (en) | 2018-04-19 | 2018-04-19 | User gender prediction method, apparatus, storage medium and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108564220A true CN108564220A (en) | 2018-09-21 |
Family
ID=63535967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810354699.XA Pending CN108564220A (en) | 2018-04-19 | 2018-04-19 | User gender prediction method, apparatus, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108564220A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241090A (en) * | 2018-09-29 | 2019-01-18 | 北京惠每云科技有限公司 | A kind of user determines method and apparatus |
| CN109933727A (en) * | 2019-03-22 | 2019-06-25 | 北京博明信德科技有限公司 | User's portrait generation method and system, user's portrait application method and system |
| CN111178983A (en) * | 2020-01-03 | 2020-05-19 | 北京搜狐新媒体信息技术有限公司 | User gender prediction method, device, equipment and storage medium |
| CN112541010A (en) * | 2019-09-23 | 2021-03-23 | 银橙(上海)信息技术有限公司 | User gender prediction method based on logistic regression |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104991899A (en) * | 2015-06-02 | 2015-10-21 | 广州酷狗计算机科技有限公司 | Identification method and apparatus of user property |
| CN105069041A (en) * | 2015-07-23 | 2015-11-18 | 合一信息技术（北京）有限公司 | Video user gender classification based advertisement putting method |
| CN105095401A (en) * | 2015-07-07 | 2015-11-25 | 北京嘀嘀无限科技发展有限公司 | Method and apparatus for identifying gender |
| CN105933425A (en) * | 2016-05-18 | 2016-09-07 | 北京奇虎科技有限公司 | Application recommendation method and device |
| CN106682686A (en) * | 2016-12-09 | 2017-05-17 | 北京拓明科技有限公司 | User gender prediction method based on mobile phone Internet-surfing behavior |
| CN106778843A (en) * | 2016-11-30 | 2017-05-31 | 腾云天宇科技（北京）有限公司 | One kind prediction mobile terminal user's property method for distinguishing, server and system |
| CN107766891A (en) * | 2017-10-31 | 2018-03-06 | 广东欧珀移动通信有限公司 | User's gender identification method, device, storage medium and electronic equipment |
-
2018
- 2018-04-19 CN CN201810354699.XA patent/CN108564220A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104991899A (en) * | 2015-06-02 | 2015-10-21 | 广州酷狗计算机科技有限公司 | Identification method and apparatus of user property |
| CN105095401A (en) * | 2015-07-07 | 2015-11-25 | 北京嘀嘀无限科技发展有限公司 | Method and apparatus for identifying gender |
| CN105069041A (en) * | 2015-07-23 | 2015-11-18 | 合一信息技术（北京）有限公司 | Video user gender classification based advertisement putting method |
| CN105933425A (en) * | 2016-05-18 | 2016-09-07 | 北京奇虎科技有限公司 | Application recommendation method and device |
| CN106778843A (en) * | 2016-11-30 | 2017-05-31 | 腾云天宇科技（北京）有限公司 | One kind prediction mobile terminal user's property method for distinguishing, server and system |
| CN106682686A (en) * | 2016-12-09 | 2017-05-17 | 北京拓明科技有限公司 | User gender prediction method based on mobile phone Internet-surfing behavior |
| CN107766891A (en) * | 2017-10-31 | 2018-03-06 | 广东欧珀移动通信有限公司 | User's gender identification method, device, storage medium and electronic equipment |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241090A (en) * | 2018-09-29 | 2019-01-18 | 北京惠每云科技有限公司 | A kind of user determines method and apparatus |
| CN109933727A (en) * | 2019-03-22 | 2019-06-25 | 北京博明信德科技有限公司 | User's portrait generation method and system, user's portrait application method and system |
| CN112541010A (en) * | 2019-09-23 | 2021-03-23 | 银橙(上海)信息技术有限公司 | User gender prediction method based on logistic regression |
| CN112541010B (en) * | 2019-09-23 | 2023-05-23 | 银橙(上海)信息技术有限公司 | User gender prediction method based on logistic regression |
| CN111178983A (en) * | 2020-01-03 | 2020-05-19 | 北京搜狐新媒体信息技术有限公司 | User gender prediction method, device, equipment and storage medium |
| CN111178983B (en) * | 2020-01-03 | 2024-03-12 | 北京搜狐新媒体信息技术有限公司 | User gender prediction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220391773A1 (en) | Method and system for artificial intelligence learning using messaging service and method and system for relaying answer using artificial intelligence | |
| CN108564220A (en) | User gender prediction method, apparatus, storage medium and computer equipment | |
| CN105630977B (en) | Application program recommended method, apparatus and system | |
| CN107729532A (en) | A kind of resume matching process and computing device | |
| CN107103036B (en) | Method and equipment for acquiring application downloading probability and programmable equipment | |
| CN109299316B (en) | Music recommendation method and device and computer equipment | |
| US20210019375A1 (en) | Computing system including virtual agent bot providing semantic topic model-based response | |
| CN108769125B (en) | Application recommendation method and device, storage medium and computer equipment | |
| CN111552880A (en) | Data processing method, device and medium based on knowledge graph and electronic equipment | |
| CN111275205B (en) | Virtual sample generation method, terminal equipment and storage medium | |
| CN107046498A (en) | The method and system of message function in expanding communication session | |
| CN111429204A (en) | Hotel recommendation method, system, electronic equipment and storage medium | |
| CN107273979A (en) | The method and system of machine learning prediction are performed based on service class | |
| CN108512883A (en) | A kind of information-pushing method, device and readable medium | |
| CN111460290A (en) | Information recommendation method, device, equipment and storage medium | |
| CN112182281B (en) | Audio recommendation method, device and storage medium | |
| CN107944026A (en) | A kind of method, apparatus, server and the storage medium of atlas personalized recommendation | |
| US20110302174A1 (en) | Crowd-sourcing for gap filling in social networks | |
| CN114040216A (en) | Live broadcast room recommendation method, medium, device and computing equipment | |
| CN113869377A (en) | Training method and device and electronic equipment | |
| CN112269943B (en) | Information recommendation system and method | |
| CN117194772B (en) | Content pushing method and device based on user tag | |
| CN112418442A (en) | Data processing method, device, equipment and storage medium for federal transfer learning | |
| CN112381236A (en) | Data processing method, device, equipment and storage medium for federal transfer learning | |
| US20140178848A1 (en) | Method and apparatus for administering learning contents |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200424 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 510640 Guangdong city of Guangzhou province Whampoa Tianhe District Road No. 163 Xiping Yun Lu Yun Ping square B radio tower 15 layer self unit 02 Applicant before: GUANGZHOU UC NETWORK TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180921 |
|
| WD01 | Invention patent application deemed withdrawn after publication |