CN109495552A - Method and apparatus for updating clicking rate prediction model - Google Patents
Method and apparatus for updating clicking rate prediction model Download PDFInfo
- Publication number
- CN109495552A CN109495552A CN201811288021.2A CN201811288021A CN109495552A CN 109495552 A CN109495552 A CN 109495552A CN 201811288021 A CN201811288021 A CN 201811288021A CN 109495552 A CN109495552 A CN 109495552A
- Authority
- CN
- China
- Prior art keywords
- information
- clicking rate
- displaying information
- prediction model
- displaying
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Abstract
The embodiment of the present application discloses the method and apparatus for updating clicking rate prediction model.One specific embodiment of this method includes: to receive at least one to show information and display;Obtain the user information of target user;Displaying information in information is shown at least one, obtains the characteristic information of the displaying information;In response to detecting that target user is directed to the object run that the displaying information carries out, markup information corresponding with object run is generated;Utilize machine learning method, the characteristic information and user information that at least one is shown the displaying information in information are as the input of initial clicking rate prediction model, using markup information corresponding with the characteristic information of input and user information as the desired output of initial clicking rate prediction model, training obtains updated clicking rate prediction model.The embodiment realizes update clicking rate prediction model in time, helps to improve the accuracy that clicking rate prediction is carried out using the model that training obtains.
Description
Technical field
The invention relates to field of computer technology, and in particular to for update clicking rate prediction model method and
Device.
Background technique
Clicking rate is also known as click-through-rate (CTR, Click Through Rate), i.e. certain information is actually clicked number
Divided by the result of the displaying amount (such as quantity of the terminal of the number or reception information of pushed information) of the information.According to click
Rate can reflect the effect of information displaying, and alternatively information, pushed information and the reference being ranked up etc. to information, from
And the specific aim of the operations such as information push, displaying can be improved.Using existing for predicting the model of clicking rate, can predict
The clicking rate of information can treat the information of push using the clicking rate of prediction or information to be shown is targetedly selected
It selects or sorts.
Summary of the invention
The embodiment of the present application proposes the method and apparatus for updating clicking rate prediction model, and for showing information
Method and apparatus.
In a first aspect, the embodiment of the present application provide it is a kind of for updating the method for clicking rate prediction model, this method packet
It includes: receiving at least one and show information and display;Obtain the user information of target user;At least one is shown in information
It shows information, obtains the characteristic information of the displaying information;In response to detecting that target user is directed to the mesh that the displaying information carries out
Mark operation, generates markup information corresponding with object run;Using machine learning method, at least one is shown into the exhibition in information
Show the input of the characteristic information and user information of information as initial clicking rate prediction model, by the characteristic information and use with input
Desired output of the information corresponding markup information in family as initial clicking rate prediction model, it is pre- that training obtains updated clicking rate
Survey model.
In some embodiments, object run is that clicking operation or click operate;And in response to detecting that target is used
Family is directed to the object run that the displaying information carries out, and generates markup information corresponding with object run, comprising: in response to detecting
Target user generates the first information corresponding with clicking operation as markup information to the clicking operation of the displaying information;Response
In detecting the click operation for characterizing target user and not clicking on the displaying information, the second letter corresponding with click operation is generated
Breath is used as markup information.
In some embodiments, click operation, which includes at least one of the following:, hides operation, delete operation, wherein hides
Operation is in the case where showing that information is not clicked on, to show that information is moved out of the operation of preset display area.
In some embodiments, after the characteristic information for obtaining the displaying information, this method further include: in response to detection
Into the preset duration after the time for receiving the displaying information, target user does not click on the displaying information, generates characterization
The markup information that the displaying information is not clicked on.
Second aspect, the embodiment of the present application provide a kind of method for showing information, this method comprises: receiving at least
Two displaying information;Obtain the user information of target user;For the displaying information at least two displaying information, the exhibition is obtained
Show the characteristic information of information;By acquired characteristic information and user information input clicking rate prediction model trained in advance, obtain
To prediction clicking rate, wherein clicking rate prediction model is according to claim 1 one of -4 obtained updated click of method
Rate prediction model;According to the size of obtained prediction clicking rate, the displaying information at least two displaying information is arranged
At least two after sequence, and display sequence show the displaying information in information.
The third aspect, the embodiment of the present application provide a kind of for updating the device of clicking rate prediction model, the device packet
Include: receiving unit is configured to receive at least one and shows information and display;Acquiring unit is configured to obtain target user
User information;Generation unit is configured to show the displaying information in information at least one, obtains the displaying information
Characteristic information;In response to detecting that target user is directed to the object run that the displaying information carries out, generate corresponding with object run
Markup information;Training unit, is configured to using machine learning method, at least one is shown the displaying information in information
The input of characteristic information and user information as initial clicking rate prediction model, by the characteristic information and user information pair with input
Desired output of the markup information answered as initial clicking rate prediction model, training obtain updated clicking rate prediction model.
In some embodiments, object run is that clicking operation or click operate;And generation unit includes: first raw
At module, it is configured in response to detect that target user to the clicking operation of the displaying information, generates corresponding with clicking operation
The first information as markup information;Second generation module is configured in response to detect that characterization target user does not click on this
The click operation for showing information, generates the second information corresponding with click operation as markup information.
In some embodiments, click operation, which includes at least one of the following:, hides operation, delete operation, wherein hides
Operation is in the case where showing that information is not clicked on, to show that information is moved out of the operation of preset display area.
In some embodiments, generation unit is further configured to: receiving the displaying information in response to detecting
Time after preset duration in, target user does not click on the displaying information, generates and characterizes what the displaying information was not clicked on
Markup information.
Fourth aspect, the embodiment of the present application provide it is a kind of for showing the device of information, the device include: receive it is single
Member is configured to receive at least two displaying information;Acquiring unit is configured to obtain the user information of target user;Prediction
Unit is configured to obtain the characteristic information of the displaying information for the displaying information at least two displaying information;It will be obtained
Characteristic information and the user information input taken clicking rate prediction model trained in advance, obtains prediction clicking rate, wherein clicking rate
Prediction model is that the updated clicking rate obtained according to the method for any embodiment description in above-mentioned first aspect predicts mould
Type;Display unit is configured to the size according to obtained prediction clicking rate, shows that the displaying in information is believed at least two
Breath is ranked up, and at least two after display sequence show the displaying information in information.
5th aspect, the embodiment of the present application provide a kind of terminal device, which includes: one or more processing
Device;Storage device is stored thereon with one or more programs;When one or more programs are executed by one or more processors,
So that one or more processors realize the method as described in implementation any in first aspect or second aspect.
6th aspect, the embodiment of the present application provide a kind of computer-readable medium, are stored thereon with computer program, should
The method as described in implementation any in first aspect or second aspect is realized when computer program is executed by processor.
The method and apparatus provided by the embodiments of the present application for being used to update clicking rate prediction model, by receiving at least one
It shows information and display, and obtains the user information of target user, obtain the characteristic information for showing information, detect target user
Operation to information is shown generates the corresponding mark of object run if user has carried out object run for displaying information
Information, using the characteristic information for showing information, user information markup information corresponding with information is shown as training sample, to initial
Clicking rate prediction model is trained, the clicking rate prediction model after being trained, to realize real-time monitoring target user
To the operation for showing information and generate training sample in real time, using the training sample progress model training generated in real time, in time
Clicking rate prediction model is updated, the accuracy for carrying out clicking rate prediction using the model that training obtains is helped to improve.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that one embodiment of the application can be applied to exemplary system architecture figure therein;
Fig. 2 is the process according to one embodiment of the method for updating clicking rate prediction model of the embodiment of the present application
Figure;
Fig. 3 is showing for application scenarios of the method according to the embodiment of the present application for updating clicking rate prediction model
It is intended to;
Fig. 4 is the flow chart according to one embodiment of the method for the display information of the embodiment of the present application;
Fig. 5 is the structure according to one embodiment of the device for updating clicking rate prediction model of the embodiment of the present application
Schematic diagram;
Fig. 6 is the structural schematic diagram according to one embodiment of the device for showing information of the embodiment of the present application;
Fig. 7 is adapted for the structural schematic diagram for the computer system for realizing the terminal device of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the embodiment of the present application for updating the method for clicking rate prediction model or for updating
The device of clicking rate prediction model, the method for showing information, the exemplary system architecture of the device for showing information
100。
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed, such as web browser is answered on terminal device 101,102,103
With, shopping class application, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard
When part, it can be the various electronic equipments with display screen, including but not limited to smart phone, tablet computer, on knee portable
Computer and desktop computer etc..When terminal device 101,102,103 is software, above-mentioned cited electricity may be mounted at
In sub- equipment.Multiple softwares or software module may be implemented into (such as providing the software of Distributed Services or software mould in it
Block), single software or software module also may be implemented into.It is not specifically limited herein.
Server 105 can be to provide the server of various services, such as to showing on terminal device 101,102,103
Show that information provides the background information push server supported.Background information push server can be pushed to terminal device and be shown
Information, to show on the terminal device.Terminal device, which can use, shows that information is trained initial model, after obtaining update
Model.
It should be noted that for updating the method for clicking rate prediction model generally by end provided by the embodiment of the present application
End equipment 101,102,103 executes, and correspondingly, the device for updating clicking rate prediction model is generally positioned at terminal device
101, in 102,103.For showing the method for information generally by terminal device 101,102,103 provided by the embodiment of the present application
It executes, correspondingly, for showing that the device of information is generally positioned in terminal device 101,102,103.
It should be noted that server can be hardware, it is also possible to software.When server is hardware, may be implemented
At the distributed server cluster that multiple servers form, individual server also may be implemented into.It, can when server is software
To be implemented as multiple softwares or software module (such as providing the software of Distributed Services or software module), also may be implemented
At single software or software module.It is not specifically limited herein.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, an implementation of the method for updating clicking rate prediction model according to the application is shown
The process 200 of example.This is used to update the method for clicking rate prediction model, comprising the following steps:
Step 201, it receives at least one and shows information and display.
In the present embodiment, for updating executing subject (such as the terminal shown in FIG. 1 of the method for clicking rate prediction model
Equipment) it can be by wired connection mode or radio connection from for pushing server (such as Fig. 1 institute for showing information
The server shown) at least one displaying information is received, and at least one received displaying information is shown in above-mentioned execution master
On the display screen of body.It is at least one of following to show that information can include but is not limited to: picture, text, audio, video, chain ground connection
Location etc..In general, showing that information can be the information for characterizing the information such as article, video, audio;Alternatively, showing information
It is also possible to characterize the information of product, product can be physical product, be also possible to virtual product (such as software, file etc.).
When showing that information is shown in above-mentioned executing subject, the operations such as user can click, collect to displaying information.
Step 202, the user information of target user is obtained.
In the present embodiment, above-mentioned executing subject can be from user information long-range or from local acquisition target user.Its
In, target user can be the user that information is shown using above-mentioned executing subject browsing, and the user information of target user can be pre-
Be first stored in other electronic equipments in above-mentioned executing subject or with the communication connection of above-mentioned executing subject (such as it is shown in FIG. 1
Server).The user information of target user can be used for characterizing the feature of target user, and the feature of target user includes but unlimited
In following at least one: gender, age, interest of target user etc..
Step 203, the displaying information in information is shown at least one, obtains the characteristic information of the displaying information;It rings
Ying Yu detects that target user is directed to the object run that the displaying information carries out, and generates markup information corresponding with object run.
In the present embodiment, for the displaying information at least one displaying information received in step 201, above-mentioned execution
Main body can execute following steps:
Firstly, obtaining the characteristic information of the displaying information.Wherein, characteristic information is used to characterize the feature for showing information.It is special
Reference breath may include be contained in show information in information (such as show information title, show information include chain ground connection
Location etc.), above-mentioned executing subject can extract these information as characteristic information from showing in information.Characteristic information also may include
In advance for the information (such as type belonging to displaying information, the author for showing information etc.) for showing that information is established, above-mentioned execution
Main body can be from remotely or from the above-mentioned preparatory information established for displaying information of local acquisition as characteristic information.
Then, in response to detecting that target user is directed to the object run that the displaying information carries out, generation and object run
Corresponding markup information.Specifically, object run can be pre-set at least one operation, and object run may include each
Kind operation, such as clicking operation, delete operation etc..The corresponding relationship of object run and markup information can be by the table that pre-establishes
The characterization such as lattice, chained list.As an example, the corresponding relationship of object run and markup information can be by the mapping table that pre-establishes
It characterizes, can store the mark and the corresponding markup information of each mark of multiple object runs in the mapping table.On
The mark of object run can be determined according to the object run detected by stating executing subject, from the mapping table search with
The corresponding markup information of the mark is as the corresponding markup information of object run detected.Markup information can be used for characterizing mesh
Whether mark user clicks the displaying information, and markup information may include but be limited to the information of following at least one form: number, text
Word, symbol etc..For example, markup information can be digital " 1 " or " 0 ", wherein " 1 " indicates to click, and " 0 " indicates not clicking on.
In some optional implementations of the present embodiment, the object run that above-mentioned executing subject detects can be point
Hit operation or click operation.Above-mentioned executing subject can be in response to detecting that target user grasps the click of the displaying information
Make, generates the preset first information corresponding with clicking operation as markup information.As an example, the first information can be number
"1".In general, as long as above-mentioned executing subject detects clicking operation, i.e. the generation first information.
Above-mentioned executing subject can in response to detect characterization target user do not click on the displaying information click operate,
Preset the second information corresponding with click operation is generated as markup information.As an example, the second information can be number
“0”。
In some optional implementations of the present embodiment, click operation may include at least one of following: hide
Operation, delete operation etc..Wherein, hiding operation can be show that information is not clicked in the case where, it is pre- to show that information is moved out of
If display area operation.As an example, when above-mentioned executing subject includes the display screen with touch function, if detection
Show that information is not clicked to certain, and the displaying information is removed display area by upper cunning or operation of gliding by target user,
Then determination performs hiding operation at this time.Above-mentioned preset display area can be for showing the region for showing information, this is aobvious
Show that region can be whole region or the partial region of the display screen of above-mentioned executing subject.In general, if target user will show
Information removes display area, illustrates that target user is little to the interest of the displaying information, at this point it is possible to generate the second information conduct
Markup information.
In addition, above-mentioned delete operation can be in the case where showing that information is not clicked on, the deleted operation of information is shown.
In general, target user can carry out delete operation by clicking delete button.
In some optional implementations of the present embodiment, above-mentioned executing subject may also respond to detect and receive
In preset duration after to the time of the displaying information, target user does not click on the displaying information, generates and characterizes displaying letter
Cease the markup information being not clicked on.As an example it is supposed that preset duration is 10 minutes, then above-mentioned executing subject can be from receiving
Start timing at the time of the displaying information, information is shown for some, if the click is not shown target user in 10 minutes,
It then generates and characterizes the markup information (such as digital " 0 ") that the displaying information is not clicked on.By this implementation, can make
It states executing subject and generates the corresponding markup information of each displaying information within a certain period of time, help to improve the effect of model training
Rate updates clicking rate prediction model in time.
Step 204, using machine learning method, at least one is shown to the characteristic information and use of the displaying information in information
Input of the family information as initial clicking rate prediction model, by markup information corresponding with the characteristic information of input and user information
As the desired output of initial clicking rate prediction model, training obtains updated clicking rate prediction model.
In the present embodiment, above-mentioned executing subject can use machine learning method, at least one is shown in information
The input as initial clicking rate prediction model of characteristic information and user information for showing information, by with the characteristic information of input and
Desired output of the corresponding markup information of user information as initial clicking rate prediction model, training obtain updated clicking rate
Prediction model.
Specifically, above-mentioned initial clicking rate prediction model can be unbred model, be also possible to trained mistake
Model.Above-mentioned initial clicking rate prediction model can include but is not limited to following at least one model: FM
(Factorization Machine, Factorization machine) model, FFM (Field-aware Factorization Machine,
Field perception Factorization machine, neural network model etc..Initial parameter has can be set in initial clicking rate prediction model, and parameter is being instructed
It can be continuously adjusted during white silk.Herein, when initial point, which hits rate prediction model, meets at least one of following condition, really
Fixed training is completed, and updated clicking rate prediction model is obtained: it is above-mentioned at least one show each displaying information in information
Characteristic information and corresponding markup information are used to have trained initial clicking rate prediction model;Resulting difference is calculated to be less than in advance
If discrepancy threshold.
Wherein, above-mentioned executing subject can calculate penalty values based on preset loss function, and penalty values can be used for characterizing
Difference between reality output and desired output.In practice, reality output phase can be calculated using preset various loss functions
For the penalty values of the output of mark.For example, penalty values can be calculated using logarithm loss function, cross entropy loss function etc..
In general, clicking rate prediction model can be the model for classifying to the one of the various information of input.For example,
When markup information is " 0 ", characterization user does not click on displaying information, that is, inputted when training with letter corresponding to the markup information
Breath belongs to click class;When markup information is " 1 ", characterization user clicks displaying information, that is, inputted when training with the mark
Information corresponding to note information belongs to click class.Clicking rate prediction model after training can be with output probability value, and probability value can be with
Various information for characterizing input belong to click class or click the probability of class, the various information of characterization input can be belonged to
The probability of class is clicked as prediction clicking rate.It predicts that clicking rate is bigger, indicates to show that a possibility that information is clicked by user is bigger.
With continued reference to the application scenarios that Fig. 3, Fig. 3 are according to the method for updating clicking rate prediction model of the present embodiment
A schematic diagram.In the application scenarios of Fig. 3, terminal device 301 is firstly received multiple displaying information 302, and will be each
Show that information is shown on the screen of terminal device 301.Then, terminal device 301 obtains target user's (i.e. currently used terminal
The user of equipment 301) user information 303.Wherein, user information 303 includes gender information, locating geographical location information, year
Age information etc..
Subsequently, for above-mentioned multiple each displaying information shown in information 302, terminal device 301 obtains the displaying
The characteristic information of information;Object run (such as the clicking operation carried out in response to detecting target user for the displaying information
Or click operation), generation markup information corresponding with object run (such as " 1 " of characterization clicking operation, characterization click behaviour
" 0 " made).Wherein, clicking operation is the operation that target user clicks the displaying information, and click operation is that target is used
The displaying information is skidded off the operation of the display area of the screen of terminal device 301 by family.
Finally, terminal device 301 utilizes machine learning method, above-mentioned multiple each displayings shown in information 302 are believed
The characteristic information (such as 306 in figure are that some shows the corresponding characteristic information of information) and user information 303 of breath are as initial
The input of clicking rate prediction model 304 (such as schemes markup information corresponding with the characteristic information of input and user information 303
In 307 show the corresponding markup information of information for some) desired output as initial clicking rate prediction model 304, it is trained
Obtain updated clicking rate prediction model 305.
The method provided by the above embodiment of the application shows information and display by receiving at least one, and obtains
The user information of target user obtains the characteristic information for showing information, detects operation of the target user to information is shown, if with
Family has carried out object run for displaying information, then generates the corresponding markup information of object run, will show that the feature of information is believed
Breath, user information markup information corresponding with information is shown are trained initial clicking rate prediction model as training sample,
Clicking rate prediction model after being trained, to realize real-time monitoring target user to the operation for showing information and give birth in real time
At training sample, model training is carried out using the training sample generated in real time, clicking rate prediction model is updated in time, facilitates
Improve the accuracy that clicking rate prediction is carried out using the model that training obtains.
With further reference to Fig. 4, it illustrates the processes 400 of one embodiment for showing the method for information.This is used for
Show the process 400 of the method for information, comprising the following steps:
Step 401, at least two are received and shows information.
In the present embodiment, for showing that the executing subject (such as terminal device shown in FIG. 1) of the method for information can be with
By wired connection mode or radio connection from for pushing server (such as the service shown in FIG. 1 for showing information
Device) receive at least two displaying information.Wherein, for showing that the description of information may refer to the step of above-mentioned Fig. 2 corresponding embodiment
Rapid 201, which is not described herein again.
Step 402, the user information of target user is obtained.
In the present embodiment, the user information of the available target user of above-mentioned executing subject.Wherein, target user can be with
It is that the user for showing information is browsed using above-mentioned executing subject, the user information of target user can be stored in advance in above-mentioned execution
In other electronic equipments in main body or with the communication connection of above-mentioned executing subject (such as server shown in FIG. 1).Target user
User information can be used for characterizing the feature of target user, the feature of target user includes but is not limited to following at least one:
Gender, age, interest of target user etc..
Step 403, for the displaying information at least two displaying information, the characteristic information of the displaying information is obtained;It will
The clicking rate prediction model that acquired characteristic information and user information input are trained in advance, obtains prediction clicking rate.
In the present embodiment, received for step 401 at least two the displaying information in information, above-mentioned execution master are shown
Body can execute following steps:
Firstly, obtaining the characteristic information of the displaying information.Wherein, the description as described in characteristic information may refer to above-mentioned Fig. 2
The step 203 of corresponding embodiment, which is not described herein again.
Then, the clicking rate prediction model that acquired characteristic information and user information input is trained in advance, obtains pre-
Survey clicking rate.Wherein, the description as described in prediction clicking rate may refer to the description in above-mentioned Fig. 2 embodiment.Clicking rate predicts mould
Type can be the updated clicking rate prediction model generated using the method as described in above-mentioned Fig. 2 embodiment.It is specific raw
It may refer to the associated description of Fig. 2 embodiment at process, details are not described herein.
It should be noted that the executing subject of the present embodiment can be identical as the executing subject of above-mentioned Fig. 2 corresponding embodiment
Or it is different.When identical, executed the step of training pattern and using the step of model in same terminal device.In this way, above-mentioned hold
Row main body can carry out model training, obtained update based on the displaying information received after receiving displaying information every time
Clicking rate prediction model afterwards.After receiving displaying information next time, it is pre- that updated clicking rate prediction model can be used
The clicking rate for showing information is surveyed, and using above-mentioned updated clicking rate prediction model as initial clicking rate prediction model, base
Continue model training in the displaying information that this is received, obtains updated clicking rate prediction model again.
Step 404, according to it is obtained prediction clicking rate size, at least two show information in displaying information into
At least two after row sequence, and display sequence show the displaying information in information.
In the present embodiment, above-mentioned executing subject can be according to the size of obtained prediction clicking rate, at least two
Show that the displaying information in information is ranked up, and at least two after display sequence show the displaying information in information.
In general, above-mentioned executing subject can be according to the descending sequence of prediction clicking rate to carrying out in each displaying information
Sequence, according to each displaying information of sequencing display.Such as the biggish displaying information of corresponding clicking rate is shown in display area
Top, in this way, family, which can be used, preferentially clicks the corresponding biggish displaying information of clicking rate, thus targetedly to user
Show each displaying information.
The method provided by the above embodiment of the application, trained by using the method described by above-mentioned Fig. 2 embodiment
The updated clicking rate prediction model arrived, the corresponding prediction clicking rate of each displaying information received, according to each
The size of clicking rate is predicted to each displaying information sorting and display, shows each specific aim for showing information to improve,
It enriches and shows each mode for showing information.
With further reference to Fig. 5, as the realization to method shown in above-mentioned Fig. 2, this application provides one kind for updating a little
One embodiment of the device of rate prediction model is hit, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, the device
It specifically can be applied in various electronic equipments.
As shown in figure 5, the present embodiment includes: receiving unit 501 for updating the device 500 of clicking rate prediction model,
It is configured to receive at least one and shows information and display;Acquiring unit 502 is configured to obtain user's letter of target user
Breath;Generation unit 503 is configured to show the displaying information in information at least one, obtains the feature of the displaying information
Information;In response to detecting that target user is directed to the object run that the displaying information carries out, mark corresponding with object run is generated
Infuse information;Training unit 504, is configured to using machine learning method, at least one is shown the displaying information in information
The input of characteristic information and user information as initial clicking rate prediction model, by the characteristic information and user information pair with input
Desired output of the markup information answered as initial clicking rate prediction model, training obtain updated clicking rate prediction model.
In the present embodiment, receiving unit 501 can be by wired connection mode or radio connection from for pushing away
Putting on display show the server (such as server shown in FIG. 1) of information receive at least one show information, and by it is received at least
One displaying information is shown on the display screen of above-mentioned apparatus 500.It is at least one of following to show that information can include but is not limited to:
Picture, text, audio, video, chained address etc..In general, showing that information can be for characterizing such as article, video, audio
Etc. information information;Alternatively, showing that information is also possible to characterize the information of product, product can be physical product, be also possible to
Virtual product (such as software, file etc.).When showing that information is shown in above-mentioned apparatus 500, user can be to displaying
Information such as is clicked, is collected at the operation.
In the present embodiment, acquiring unit 502 can be from user information long-range or from local acquisition target user.Its
In, target user can be the user for browsing using above-mentioned apparatus 500 and showing information, and the user information of target user can be preparatory
Be stored in other electronic equipments in above-mentioned apparatus 500 or with the communication connection of above-mentioned apparatus 500 (such as service shown in FIG. 1
Device).The user information of target user can be used for characterizing the feature of target user, the feature of target user include but is not limited to
Lower at least one: gender, age, interest of target user etc..
In the present embodiment, the displaying information at least one displaying information received for receiving unit 501, above-mentioned life
Following steps can be executed at unit 503:
Firstly, obtaining the characteristic information of the displaying information.Wherein, characteristic information is used to characterize the feature for showing information.It is special
Reference breath may include be contained in show information in information (such as show information title, show information include chain ground connection
Location etc.), above-mentioned generation unit 503 can extract these information as characteristic information from showing in information.Characteristic information can also be with
It is above-mentioned including the information (such as type belonging to displaying information, the author for showing information etc.) established in advance for displaying information
Generation unit 503 can be from long-range or from local obtain these information as characteristic information.
Then, in response to detecting that target user is directed to the object run that the displaying information carries out, generation and object run
Corresponding markup information.Specifically, object run may include various operations, such as clicking operation, delete operation etc..Target behaviour
Make to be characterized by table, the chained list etc. pre-established with the corresponding relationship of markup information.As an example, object run and mark
The corresponding relationship of information can be characterized by the mapping table pre-established, can store multiple mesh in the mapping table
Mark the mark and the corresponding markup information of each mark of operation.Above-mentioned generation unit 503 can according to the object run detected,
The mark for determining object run searches markup information corresponding with the mark as the target detected from the mapping table
Operate corresponding markup information.Markup information can be used for characterizing target user and click the displaying information or not click on displaying letter
Breath, markup information may include but be limited to the information of following at least one form: number, text, symbol etc..For example, mark letter
Breath can be digital " 1 " or " 0 ", wherein " 1 " indicates to click, and " 0 " indicates not clicking on.
In the present embodiment, training unit 504 can use machine learning method, at least one is shown the exhibition in information
Show the input of the characteristic information and user information of information as initial clicking rate prediction model, by the characteristic information and use with input
Desired output of the information corresponding markup information in family as initial clicking rate prediction model, it is pre- that training obtains updated clicking rate
Survey model.
Specifically, above-mentioned initial clicking rate prediction model can be unbred model, be also possible to trained mistake
Model.Above-mentioned initial clicking rate prediction model can include but is not limited to following at least one model: FM
(Factorization Machine, Factorization machine) model, FFM (Field-aware Factorization Machine,
Field perception Factorization machine, neural network model etc..Initial parameter has can be set in initial clicking rate prediction model, and parameter is being instructed
It can be continuously adjusted during white silk.Herein, when initial point, which hits rate prediction model, meets at least one of following condition, really
Fixed training is completed, and updated clicking rate prediction model is obtained: it is above-mentioned at least one show each displaying information in information
Characteristic information and corresponding markup information are used to have trained initial clicking rate prediction model;Resulting difference is calculated to be less than in advance
If discrepancy threshold.
In some optional implementations of the present embodiment, object run is that clicking operation or click operate;And
Generation unit 503 may include: the first generation module (not shown), be configured in response to detect target user to this
It shows the clicking operation of information, generates the first information corresponding with clicking operation as markup information;Second generation module is (in figure
Be not shown), be configured in response to detect that characterization target user does not click on the click operation of the displaying information, generate with it is non-
Corresponding second information of clicking operation is as markup information.
In some optional implementations of the present embodiment, click operation include at least one of the following: hide operation,
Delete operation, wherein hiding operation is in the case where showing that information is not clicked on, to show that information is moved out of preset display area
Operation.
In some optional implementations of the present embodiment, generation unit 503 can be further configured to: in response to
It detects in the preset duration after the time for receiving the displaying information, target user does not click on the displaying information, generates
Characterize the markup information that the displaying information is not clicked on.
The device provided by the above embodiment of the application shows information and display by receiving at least one, and obtains
The user information of target user obtains the characteristic information for showing information, detects operation of the target user to information is shown, if with
Family has carried out object run for displaying information, then generates the corresponding markup information of object run, will show that the feature of information is believed
Breath, user information markup information corresponding with information is shown are trained initial clicking rate prediction model as training sample,
Clicking rate prediction model after being trained, to realize real-time monitoring target user to the operation for showing information and give birth in real time
At training sample, model training is carried out using the training sample generated in real time, clicking rate prediction model is updated in time, facilitates
Improve the accuracy that clicking rate prediction is carried out using the model that training obtains.
With further reference to Fig. 6, as the realization to method shown in above-mentioned Fig. 4, this application provides one kind for showing letter
One embodiment of the device of breath, the Installation practice is corresponding with embodiment of the method shown in Fig. 4, which can specifically answer
For in various electronic equipments.
As shown in fig. 6, the present embodiment is used to show that the device 600 of information includes: receiving unit 601, it is configured to connect
It receives at least two and shows information;Acquiring unit 602 is configured to obtain the user information of target user;Predicting unit 603, quilt
It is configured to obtain the characteristic information of the displaying information for the displaying information at least two displaying information;By acquired spy
The clicking rate prediction model that reference breath and user information input are trained in advance, obtains prediction clicking rate, wherein clicking rate predicts mould
Type is the updated clicking rate prediction model obtained according to the method for above-mentioned Fig. 2 corresponding embodiment description;Display unit 604,
It is configured to the size according to obtained prediction clicking rate, the displaying information at least two displaying information is ranked up,
And at least two after display sequence show the displaying information in information.
In the present embodiment, receiving unit 601 can be by wired connection mode or radio connection from for pushing away
Putting on display shows that the server (such as server shown in FIG. 1) of information receives at least two and shows information.Wherein, for showing information
Description may refer to the step 201 of above-mentioned Fig. 2 corresponding embodiment, which is not described herein again.
In the present embodiment, the user information of the available target user of acquiring unit 602.Wherein, target user can be with
It is that the user for showing information is browsed using above-mentioned apparatus 600, the user information of target user can be stored in advance in above-mentioned apparatus
In other electronic equipments communicated to connect in 600 or with above-mentioned apparatus 600 (such as server shown in FIG. 1).Target user's
User information can be used for characterizing the feature of target user, and the feature of target user includes but is not limited to following at least one: mesh
Mark gender, age, the interest etc. of user.
In the present embodiment, the displaying information in information is shown for receiving unit 601 received at least two, it is above-mentioned pre-
Following steps can be executed by surveying unit 603:
Firstly, obtaining the characteristic information of the displaying information.Wherein, the description as described in characteristic information may refer to above-mentioned Fig. 2
The step 203 of corresponding embodiment, which is not described herein again.
Then, the clicking rate prediction model that acquired characteristic information and user information input is trained in advance, obtains pre-
Survey clicking rate.Wherein, the description as described in prediction clicking rate may refer to the description in above-mentioned Fig. 2 embodiment.Clicking rate predicts mould
Type can be the updated clicking rate prediction model generated using the method as described in above-mentioned Fig. 2 embodiment.It is specific raw
It may refer to the associated description of Fig. 2 embodiment at process, details are not described herein.
In the present embodiment, display unit 604 can be opened up according to the size of obtained prediction clicking rate at least two
Show that the displaying information in information is ranked up, and at least two after display sequence show the displaying information in information.
In general, above-mentioned display unit 604 can be according to the descending sequence of prediction clicking rate in each displaying information
It is ranked up, according to each displaying information of sequencing display.Such as the biggish displaying information of corresponding clicking rate is shown and is being shown
The top in region, in this way, family, which can be used, preferentially clicks the corresponding biggish displaying information of clicking rate, thus targetedly to
User shows each displaying information.
The device provided by the above embodiment of the application, trained by using the method described by above-mentioned Fig. 2 embodiment
The updated clicking rate prediction model arrived, the corresponding prediction clicking rate of each displaying information received, according to each
The size of clicking rate is predicted to each displaying information sorting and display, shows each specific aim for showing information to improve,
It enriches and shows each mode for showing information.
Below with reference to Fig. 7, it illustrates the computer systems 700 for the terminal device for being suitable for being used to realize the embodiment of the present application
Structural schematic diagram.Terminal device shown in Fig. 7 is only an example, function to the embodiment of the present application and should not use model
Shroud carrys out any restrictions.
As shown in fig. 7, computer system 700 includes central processing unit (CPU) 701, it can be read-only according to being stored in
Program in memory (ROM) 702 or be loaded into the program in random access storage device (RAM) 703 from storage section 708 and
Execute various movements appropriate and processing.In RAM 703, also it is stored with system 700 and operates required various programs and data.
CPU 701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 705 is also connected to always
Line 704.
I/O interface 705 is connected to lower component: the importation 706 including keyboard, mouse etc.;Including liquid crystal display
And the output par, c 707 of loudspeaker etc. (LCD) etc.;Storage section 708 including hard disk etc.;And including such as LAN card, tune
The communications portion 709 of the network interface card of modulator-demodulator etc..Communications portion 709 executes mailing address via the network of such as internet
Reason.Driver 710 is also connected to I/O interface 705 as needed.Detachable media 711, such as disk, CD, magneto-optic disk, half
Conductor memory etc. is mounted on as needed on driver 710, in order to as needed from the computer program read thereon
It is mounted into storage section 708.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 709, and/or from detachable media
711 are mounted.When the computer program is executed by central processing unit (CPU) 701, limited in execution the present processes
Above-mentioned function.
It should be noted that computer-readable medium described herein can be computer-readable signal media or meter
Calculation machine readable medium either the two any combination.Computer-readable medium for example may be-but not limited to-
Electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.It is computer-readable
The more specific example of medium can include but is not limited to: have electrical connection, the portable computer magnetic of one or more conducting wires
Disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or sudden strain of a muscle
Deposit), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned appoint
The suitable combination of meaning.In this application, computer-readable medium can be any tangible medium for including or store program, the journey
Sequence can be commanded execution system, device or device use or in connection.And in this application, it is computer-readable
Signal media may include in a base band or as carrier wave a part propagate data-signal, wherein carrying computer can
The program code of reading.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal, optical signal or
Above-mentioned any appropriate combination.Computer-readable signal media can also be any calculating other than computer-readable medium
Machine readable medium, the computer-readable medium can be sent, propagated or transmitted for by instruction execution system, device or device
Part uses or program in connection.The program code for including on computer-readable medium can use any Jie appropriate
Matter transmission, including but not limited to: wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The calculating of the operation for executing the application can be write with one or more programming languages or combinations thereof
Machine program code, described program design language include object oriented program language-such as Java, Smalltalk, C+
+, it further include conventional procedural programming language-such as " C " language or similar programming language.Program code can
Fully to execute, partly execute on the user computer on the user computer, be executed as an independent software package,
Part executes on the remote computer or executes on a remote computer or server completely on the user computer for part.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include receiving unit, acquiring unit, generation unit and training unit.Wherein, the title of these units not structure under certain conditions
The restriction of the pairs of unit itself, for example, receiving unit is also described as " receiving at least one and showing information and display
Unit ".
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in terminal device described in above-described embodiment;It is also possible to individualism, and without in the supplying terminal device.
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are held by the terminal device
When row, so that the terminal device: receiving at least one and show information and display;Obtain the user information of target user;For extremely
Few displaying information shown in information, obtains the characteristic information of the displaying information;In response to detecting that target user is directed to
The object run that the displaying information carries out generates markup information corresponding with object run;It, will at least using machine learning method
One shows the input as initial clicking rate prediction model of characteristic information and user information of the displaying information in information, will be with
Desired output of the corresponding markup information of characteristic information and user information of input as initial clicking rate prediction model, trained
To updated clicking rate prediction model.
In addition, when said one or multiple programs are executed by the terminal device, it is also possible that the terminal device: connecing
It receives at least two and shows information;Obtain the user information of target user;For the displaying information at least two displaying information, obtain
Take the characteristic information of the displaying information;Acquired characteristic information and user information input clicking rate trained in advance are predicted into mould
Type obtains prediction clicking rate;According to the size of obtained prediction clicking rate, the displaying information in information is shown at least two
It is ranked up, and at least two after display sequence show the displaying information in information.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (12)
1. a kind of for updating the method for clicking rate prediction model, comprising:
It receives at least one and shows information and display;
Obtain the user information of target user;
For the displaying information at least one described displaying information, the characteristic information of the displaying information is obtained;In response to detection
It is directed to the object run that the displaying information carries out to the target user, generates markup information corresponding with object run;
Using machine learning method, the characteristic information of the displaying information at least one described displaying information and the user are believed
Cease input as initial clicking rate prediction model, using markup information corresponding with the characteristic information of input and user information as
The desired output of the initial clicking rate prediction model, training obtain updated clicking rate prediction model.
2. according to the method described in claim 1, wherein, the object run is that clicking operation or click operate;And
The object run carried out in response to detecting the target user for the displaying information, generates and object run pair
The markup information answered, comprising:
In response to detecting that the target user to the clicking operation of the displaying information, generates the first letter corresponding with clicking operation
Breath is used as markup information;
In response to detecting the click operation for characterizing the target user and not clicking on the displaying information, generate and the click
Corresponding second information is operated as markup information.
3. according to the method described in claim 2, wherein, the click operation, which includes at least one of the following:, hides operation, deletes
Except operation, wherein the hiding operation is in the case where showing that information is not clicked on, to show that information is moved out of preset viewing area
The operation in domain.
4. method described in one of -3 according to claim 1, wherein after the characteristic information for obtaining the displaying information,
The method also includes:
In response to detecting in the preset duration after the time for receiving the displaying information, the target user does not click on this
It shows information, generates and characterize the markup information that the displaying information is not clicked on.
5. a kind of method for showing information, comprising:
It receives at least two and shows information;
Obtain the user information of target user;
For the displaying information in at least two displayings information, the characteristic information of the displaying information is obtained;It will be acquired
The clicking rate prediction model that characteristic information and user information input are trained in advance, obtains prediction clicking rate, wherein the point
The rate prediction model of hitting is the updated clicking rate prediction model that method described in one of -4 obtains according to claim 1;
According to the size of obtained prediction clicking rate, the displaying information in at least two displayings information is ranked up,
And described at least two after display sequence show the displaying information in information.
6. a kind of for updating the device of clicking rate prediction model, comprising:
Receiving unit is configured to receive at least one and shows information and display;
Acquiring unit is configured to obtain the user information of target user;
Generation unit is configured to obtain the spy of the displaying information for the displaying information at least one described displaying information
Reference breath;In response to detecting that the target user is directed to the object run that the displaying information carries out, generate and object run pair
The markup information answered;
Training unit is configured to using machine learning method, by the spy of the displaying information at least one described displaying information
Reference breath and input of the user information as initial clicking rate prediction model, by the characteristic information and user information with input
Desired output of the corresponding markup information as the initial clicking rate prediction model, training obtain updated clicking rate prediction
Model.
7. device according to claim 6, wherein the object run is that clicking operation or click operate;And
The generation unit includes:
First generation module is configured in response to detect that the target user to the clicking operation of the displaying information, generates
The first information corresponding with clicking operation is as markup information;
Second generation module is configured in response to detect the click that the characterization target user does not click on the displaying information
Operation generates the second information corresponding with click operation as markup information.
8. device according to claim 7, wherein the click operation, which includes at least one of the following:, hides operation, deletes
Except operation, wherein the hiding operation is in the case where showing that information is not clicked on, to show that information is moved out of preset viewing area
The operation in domain.
9. the device according to one of claim 6-8, wherein the generation unit is further configured to:
In response to detecting in the preset duration after the time for receiving the displaying information, the target user does not click on this
It shows information, generates and characterize the markup information that the displaying information is not clicked on.
10. a kind of for showing the device of information, comprising:
Receiving unit is configured to receive at least two displaying information;
Acquiring unit is configured to obtain the user information of target user;
Predicting unit is configured to obtain the spy of the displaying information for the displaying information in at least two displayings information
Reference breath;By acquired characteristic information and the user information input clicking rate prediction model trained in advance, predicted
Clicking rate, wherein the clicking rate prediction model is the updated point that method described in one of -4 obtains according to claim 1
Hit rate prediction model;
Display unit is configured to show in information according to the size of obtained prediction clicking rate to described at least two
Show that information is ranked up, and described at least two after display sequence show the displaying information in information.
11. a kind of terminal device, comprising:
One or more processors;
Storage device is stored thereon with one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 5.
12. a kind of computer-readable medium, is stored thereon with computer program, wherein the realization when program is executed by processor
Such as method as claimed in any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811288021.2A CN109495552A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for updating clicking rate prediction model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811288021.2A CN109495552A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for updating clicking rate prediction model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109495552A true CN109495552A (en) | 2019-03-19 |
Family
ID=65693419
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811288021.2A Pending CN109495552A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for updating clicking rate prediction model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109495552A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110288433A (en) * | 2019-06-12 | 2019-09-27 | 达疆网络科技(上海)有限公司 | One kind being based on clicking rate model prediction O2O real time individual sort method |
CN110489582A (en) * | 2019-08-19 | 2019-11-22 | 腾讯科技(深圳)有限公司 | Personalization shows the generation method and device, electronic equipment of image |
CN111770125A (en) * | 2019-05-23 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for pushing information |
CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
CN113051472A (en) * | 2021-03-23 | 2021-06-29 | 北京百度网讯科技有限公司 | Modeling method, device, equipment and storage medium of click through rate estimation model |
CN113780607A (en) * | 2020-11-16 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Method and device for generating model and method and device for generating information |
CN114168046A (en) * | 2021-11-05 | 2022-03-11 | 荣耀终端有限公司 | Click operation processing method, device, storage medium and computer program product |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8732222B2 (en) * | 2010-06-30 | 2014-05-20 | Microsoft Corporation | Integrating specialized knowledge sources into a general search service |
CN104767830A (en) * | 2015-04-29 | 2015-07-08 | 百度在线网络技术(北京)有限公司 | Information issuing management method and device |
CN105956146A (en) * | 2016-05-12 | 2016-09-21 | 腾讯科技(深圳)有限公司 | Article information recommending method and device |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
CN107742221A (en) * | 2016-08-23 | 2018-02-27 | 腾讯科技(深圳)有限公司 | A kind of processing method of promotion message, device and system |
-
2018
- 2018-10-31 CN CN201811288021.2A patent/CN109495552A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8732222B2 (en) * | 2010-06-30 | 2014-05-20 | Microsoft Corporation | Integrating specialized knowledge sources into a general search service |
CN104767830A (en) * | 2015-04-29 | 2015-07-08 | 百度在线网络技术(北京)有限公司 | Information issuing management method and device |
CN105956146A (en) * | 2016-05-12 | 2016-09-21 | 腾讯科技(深圳)有限公司 | Article information recommending method and device |
CN107742221A (en) * | 2016-08-23 | 2018-02-27 | 腾讯科技(深圳)有限公司 | A kind of processing method of promotion message, device and system |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
Non-Patent Citations (1)
Title |
---|
杨诚: "基于用户实时反馈的点击率预估方法", 《计算机应用》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
CN111770125A (en) * | 2019-05-23 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for pushing information |
US11962662B2 (en) | 2019-05-23 | 2024-04-16 | Beijing Wodong Tianjun Information Technology Co., Ltd. | Method and apparatus for pushing information |
CN110288433A (en) * | 2019-06-12 | 2019-09-27 | 达疆网络科技(上海)有限公司 | One kind being based on clicking rate model prediction O2O real time individual sort method |
CN110489582A (en) * | 2019-08-19 | 2019-11-22 | 腾讯科技(深圳)有限公司 | Personalization shows the generation method and device, electronic equipment of image |
CN110489582B (en) * | 2019-08-19 | 2023-11-07 | 腾讯科技(深圳)有限公司 | Method and device for generating personalized display image and electronic equipment |
CN113780607A (en) * | 2020-11-16 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Method and device for generating model and method and device for generating information |
CN113051472A (en) * | 2021-03-23 | 2021-06-29 | 北京百度网讯科技有限公司 | Modeling method, device, equipment and storage medium of click through rate estimation model |
CN114168046A (en) * | 2021-11-05 | 2022-03-11 | 荣耀终端有限公司 | Click operation processing method, device, storage medium and computer program product |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109460513A (en) | Method and apparatus for generating clicking rate prediction model | |
CN109495552A (en) | Method and apparatus for updating clicking rate prediction model | |
CN107463704A (en) | Searching method and device based on artificial intelligence | |
US20090287989A1 (en) | Presenting tag clouds based on user information | |
CN109471695A (en) | Method and apparatus for more new data | |
CN108228906B (en) | Method and apparatus for generating information | |
CN108734293A (en) | Task management system, method and apparatus | |
CN110378474A (en) | Fight sample generating method, device, electronic equipment and computer-readable medium | |
CN111125574A (en) | Method and apparatus for generating information | |
CN109815365A (en) | Method and apparatus for handling video | |
CN109976997A (en) | Test method and device | |
CN110363604A (en) | Page generation method and device | |
CN109165344A (en) | Method and apparatus for pushed information | |
CN110298716A (en) | Information-pushing method and device | |
CN108776692A (en) | Method and apparatus for handling information | |
CN109902446A (en) | Method and apparatus for generating information prediction model | |
CN110267097A (en) | Video pushing method, device and electronic equipment based on characteristic of division | |
CN109271556A (en) | Method and apparatus for output information | |
CN109933217A (en) | Method and apparatus for pushing sentence | |
CN108182472A (en) | For generating the method and apparatus of information | |
CN110278447A (en) | Video pushing method, device and electronic equipment based on continuous feature | |
CN110209658A (en) | Data cleaning method and device | |
CN107766106B (en) | The method and apparatus for generating configuration file | |
CN108959642A (en) | Method and apparatus for information to be written | |
CN108600329A (en) | For pushed information, the method and apparatus for showing information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |