CN106982250A - Information-pushing method and device - Google Patents
Information-pushing method and device Download PDFInfo
- Publication number
- CN106982250A CN106982250A CN201710125117.6A CN201710125117A CN106982250A CN 106982250 A CN106982250 A CN 106982250A CN 201710125117 A CN201710125117 A CN 201710125117A CN 106982250 A CN106982250 A CN 106982250A
- Authority
- CN
- China
- Prior art keywords
- threshold value
- user
- probability
- probability threshold
- users
- 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.)
- Granted
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The disclosure provides a kind of information-pushing method and device, and methods described includes:Destination number according to user has been pushed obtains the first probability threshold value;The corresponding target terminal of user to be pushed is determined according to first probability threshold value;Push operation is performed for the target terminal.The disclosure obtains the first probability threshold value according to the destination number for having pushed user and performs corresponding push operation, probability threshold value need not be just determined after the completion of the click probability calculation of all users and perform corresponding push operation, so as to improve the real-time of pushed information.Meanwhile, the first probability threshold value is updated by the destination number for having pushed user, the accuracy of probability threshold value can be improved, makes actually to push quantity in all users and is more nearly with predetermined push quantity.
Description
Technical field
The application is related to communication technical field, more particularly to information-pushing method and device.
Background technology
With the continuous improvement continued to develop with living standard of science and technology, intelligent terminal increasingly universalness.Intelligent terminal
Also increasing is influenceed on daily life.Personalized push can be carried out to user by intelligent terminal.For example, can be with
News information etc. is pushed to user.
At present, in order to user's progress personalized push, it is necessary to first calculate the target letter that all users treat push respectively
The click probability of breath, is then ranked up according to the order of click probability from big to small, to the forward user's pushed information that sorts.
However, due to needing to have calculated that personalized recommendation could be carried out after the click probability of all users, amount of calculation is very big, serious shadow
The real-time pushed is rung.
The content of the invention
To overcome problem present in correlation technique, present disclose provides information-pushing method and device.
According to the first aspect of the embodiment of the present disclosure there is provided a kind of information-pushing method, methods described includes:
Destination number according to user has been pushed obtains the first probability threshold value;
The corresponding target terminal of user to be pushed is determined according to first probability threshold value;
Push operation is performed for the target terminal.
Optionally, all users are divided into K parts;It is described that user couple to be pushed is determined according to first probability threshold value
The target terminal answered, including:
According to the order by 1 to K, the kth part for being not carried out push operation, k ∈ [1, K] are obtained from the K parts;
The corresponding target of user to be pushed described in being determined according to first probability threshold value from the kth certain customers is whole
End.
Optionally, the destination number is to have pushed number of users in all users, or the destination number is kth -1
Number of users has been pushed in point.
Optionally, k ∈ (1, K], the basis pushed user destination number obtain the first probability threshold value, including:
Obtain the destination number for having pushed user;
When the destination number is more than N0, the second probability threshold value of history is reconfigured, the first probability threshold value, described is obtained
One probability threshold value is more than second probability threshold value;
When the destination number is less than N0, the second probability threshold value of history is reconfigured, the first probability threshold value, described is obtained
One probability threshold value is less than second probability threshold value;
Wherein, the N0To have pushed the desired amt of user.
Optionally, if the destination number is has pushed number of users in all users,OrIf the destination number is has pushed number of users in the part of kth -1,Or
Wherein, M represents all numbers of users, MiRepresent i-th section number of users, Mk-1Represent the certain customers' number of kth -1
Amount, N represents to make a reservation for push quantity in all users.
Optionally, when the destination number is more than N0, using formula h1=h*a reconfigures the second probability threshold value of history;
When the destination number is less than N0, using formula h1=h/a reconfigures the second probability threshold value of history;
Wherein, h1First probability threshold value is represented, h represents second probability threshold value, and a represents default Dynamic gene, a
> 1.
Optionally, if k=1, the destination number that the basis has pushed user obtains the first probability threshold value, including:
The user of predetermined ratio is randomly selected from all users, sample user group is obtained;
Calculate each user in the sample user group and click on probability for the first of target information;
Determine that first is general according to push quantity is made a reservation in the described first click probability, the predetermined ratio and all users
Rate threshold value.
Optionally, described clicked on according to described first makes a reservation for push number in probability, the predetermined ratio and all users
Amount determines the first probability threshold value, including:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained;
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets;
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
Optionally, it is described determined according to first probability threshold value from the kth certain customers described in user to be pushed
Corresponding target terminal, including:
For each user in kth certain customers, obtain the user and click on probability to the second of target information;
The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, is defined as waiting to push
The corresponding target terminal of user.
According to the second aspect of the embodiment of the present disclosure there is provided a kind of information push-delivery apparatus, including:
Threshold value acquisition module, is configured as obtaining the first probability threshold value according to the destination number for having pushed user;
Terminal deciding module, is configured as determining that the corresponding target of user to be pushed is whole according to first probability threshold value
End;
Push operation module, is configured as performing push operation for the target terminal.
Optionally, all users are divided into K parts;The terminal deciding module, including:
Acquisition submodule, is configured as according to the order by 1 to K, is obtained from the K parts and is not carried out push operation
Kth part, k ∈ [1, K];
Terminal determination sub-module, is configured as determining institute from the kth certain customers according to first probability threshold value
State the corresponding target terminal of user to be pushed.
Optionally, the destination number is to have pushed number of users in all users, or the destination number is kth -1
Number of users has been pushed in point.
Optionally, k ∈ (1, K], the threshold value acquisition module includes:
Quantity acquisition submodule, is configured as obtaining the destination number for having pushed user;
Threshold value reconfigures submodule, is configured to when the destination number is more than N0, the second probability threshold value of history is reconfigured,
The first probability threshold value is obtained, first probability threshold value is more than second probability threshold value;When the destination number is less than N0, weight
The second probability threshold value of history is configured, the first probability threshold value is obtained, first probability threshold value is less than second probability threshold value;
Wherein, the N0To have pushed the desired amt of user.
Optionally, if the destination number is has pushed number of users in all users,Or
If the destination number is has pushed number of users in the part of kth -1,Or
Wherein, M represents all numbers of users, MiRepresent i-th section number of users, Mk-1Represent the certain customers' number of kth -1
Amount, N represents to make a reservation for push quantity in all users.
Optionally, the threshold value reconfigures submodule, and concrete configuration is:
When the destination number is more than N0, using formula h1=h*a reconfigures the second probability threshold value of history;
When the destination number is less than N0, using formula h1=h/a reconfigures the second probability threshold value of history;
Wherein, h1First probability threshold value is represented, h represents second probability threshold value, and a represents default Dynamic gene, a
> 1.
Optionally, if k=1, the threshold value acquisition module includes:
Customer group sampling submodule, is configured as randomly selecting the user of predetermined ratio from all users, is sampled
Customer group;
Probability calculation submodule, is configured as calculating each user in the sample user group and is directed to the first of target information
Click on probability;
Threshold value determination sub-module, is configured as clicking on probability, the predetermined ratio and all users according to described first
In make a reservation for push quantity determine the first probability threshold value.
Optionally, the threshold value determination sub-module, concrete configuration is:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained;
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets;
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
Optionally, the terminal determination sub-module, is specifically configured to:
For each user in kth certain customers, obtain the user and click on probability to the second of target information;
The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, is defined as waiting to push
The corresponding target terminal of user.
According to the third aspect of the embodiment of the present disclosure there is provided a kind of information push-delivery apparatus, including:
Processor;
Memory for storing processor-executable instruction;
Wherein, the processor is configured as:
Destination number according to user has been pushed obtains the first probability threshold value;
The corresponding target terminal of user to be pushed is determined according to first probability threshold value;
Push operation is performed for the target terminal.
The technical scheme provided by this disclosed embodiment can include the following benefits:
The disclosure obtains the first probability threshold value according to the destination number for having pushed user and performs corresponding push operation, without
Probability threshold value is just determined after the completion of the click probability calculation of all users and corresponding push operation is performed, pushed away so as to improve
Deliver letters the real-time of breath.Simultaneously as user is uneven to the click probability distribution of target information in all users, therefore pass through
The destination number for having pushed user updates the first probability threshold value, can improve the accuracy of probability threshold value, makes real in all users
Border pushes quantity and is more nearly with the predetermined quantity that pushes.
All users are divided into K parts by the disclosure, and the first probability threshold value is obtained according to the destination number for having pushed user,
Current group user is performed to push using the first probability threshold value and judged and respective handling.Handled to currently organizing all users
Afterwards, then can be according to the quantity update probability threshold value for having pushed user, under because the quantity for having pushed user changes
One group information pushes the probability threshold value in judging, so as to realize dynamic update probability threshold value, improves the accuracy of probability threshold value, enters
And make actually to push quantity in all users and be more nearly with the predetermined quantity that pushes.
The disclosure will push number of users as the influence factor for determining the first probability threshold value, Ke Yiti in all users
Height determines the accuracy of the first probability threshold value.
The disclosure will push the target of user in processed user, be compared with the desired amt that has pushed user,
According to comparative result update probability threshold value, so that the probability threshold value after being updated, realizes quick update probability threshold value.
The disclosure is by the way that probability threshold value is multiplied with the default Dynamic gene more than 1, so as to increase probability threshold value, by inciting somebody to action
Probability threshold value is divided by with the default Dynamic gene more than 1, so as to reduce probability threshold value, realizes quick update probability threshold value.
The disclosure randomly selects the user of predetermined ratio from all users, and only the first of calculating section user is clicked on general
Rate, and click on the probability threshold value of determine the probability first using first, it is possible to reduce amount of calculation.Further, since the first probability threshold value with
The correlation of target information is higher, then the first probability threshold value that this mode is obtained compared to the threshold value directly specified, accuracy compared with
It is high.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing herein is merged in specification and constitutes the part of this specification, shows the implementation for meeting the disclosure
Example, and be used to together with specification to explain the principle of the disclosure.
Fig. 1 is that a kind of information of the disclosure according to an exemplary embodiment pushes schematic diagram of a scenario.
Fig. 2 is a kind of flow chart of information-pushing method of the disclosure according to an exemplary embodiment.
Fig. 3 is the flow chart of another information-pushing method of the disclosure according to an exemplary embodiment.
Fig. 4 is a kind of block diagram of information push-delivery apparatus of the disclosure according to an exemplary embodiment.
Fig. 5 is the block diagram of another information push-delivery apparatus of the disclosure according to an exemplary embodiment.
Fig. 6 is the block diagram of another information push-delivery apparatus of the disclosure according to an exemplary embodiment.
Fig. 7 is the block diagram of another information push-delivery apparatus of the disclosure according to an exemplary embodiment.
Fig. 8 is a kind of block diagram of the device that for information pushes of the disclosure according to an exemplary embodiment.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the disclosure.
The term used in the disclosure is the purpose only merely for description specific embodiment, and is not intended to be limiting the disclosure.
" one kind ", " described " and "the" of singulative used in disclosure and the accompanying claims book are also intended to including majority
Form, unless context clearly shows that other implications.It is also understood that term "and/or" used herein refers to and wrapped
It may be combined containing one or more associated any or all of project listed.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the disclosure
A little information should not necessarily be limited by these terms.These terms are only used for same type of information being distinguished from each other out.For example, not departing from
In the case of disclosure scope, the first information can also be referred to as the second information, similarly, and the second information can also be referred to as
One information.Depending on linguistic context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determining ".
As shown in figure 1, Fig. 1, which is a kind of information of the disclosure according to an exemplary embodiment, pushes schematic diagram of a scenario.
In the scene, service end and multiple user terminals can be included.Wherein, service end can include server, server cluster or
Person's cloud platform etc..User terminal can be computer, smart mobile phone, tablet personal computer, PDA (Personal Digital
Assistant, personal digital assistant), multimedia player, the electronic equipment such as wearable device.Service end can be used to different
The user terminal at family pushes target information, for example, can be provided with millet browser in user terminal, millet browser is corresponding
Server can push news information to the user terminal for being provided with millet browser, so that user conveniently and effectively obtains news
Information.
In order to improve the clicking rate of target information and avoid the push uninterested target information of user from being brought to user
Inconvenience, can carry out personalized recommendation for different users.During personalized push, can most it feel to target information
Many individuals of interest are pushed.In correlation technique, treated respectively to realize personalized push, it is necessary to first calculate all users
The click probability of the target information of push, is then ranked up according to the order of click probability from big to small, after sequence
Probability and specified push number are clicked on, probability threshold value is determined, and to the user's pushed information for clicking on probability and being more than probability threshold value.
Wherein, all users are the objects for being possible to carry out information push, for example, it may be have registered intended application journey
The user of sequence, destination application is can to enter the program of row information push, is browsed for example, destination application can be millet
Device.The amount of all users is often very big, for example, there are up to ten million users, if having calculated ability after the click probability of all users
Personalized recommendation can be carried out, amount of calculation is very big, the time for clicking on probability consuming for calculating all users is very long, has a strong impact on
The real-time of push.
In order to avoid the defect of pushed information poor real in correlation technique, the disclosure is according to the number of targets for having pushed user
Amount obtains the first probability threshold value, and determines the corresponding target terminal of user to be pushed according to the first probability threshold value, whole for target
End performs push operation.It can be seen that, the first probability threshold value is obtained according to the destination number for having pushed user and corresponding push is performed and grasped
Make, without just determining probability threshold value after the completion of the click probability calculation of all users, and perform corresponding push operation, so that
Improve the real-time of pushed information.
Fig. 2 is a kind of flow chart of information-pushing method of the disclosure according to an exemplary embodiment, and this method can
For in service end, comprising the following steps:
In step 201, the first probability threshold value is obtained according to the destination number for having pushed user.
In step 202., the corresponding target terminal of user to be pushed is determined according to first probability threshold value.
In step 203, push operation is performed for the target terminal.
In the embodiment of the present disclosure, the user that user is executed push operation in processed user, push operation have been pushed
It can be the operation for pushing target information.Target information is information to be pushed, for example, target information can be that a physical culture is new
The information such as news, science and technology news, entertainment newses.
First probability threshold value can be for judging whether user is to need the targeted customer of progress target information push
Quantile, it is possible to use the first probability threshold value filters out user to be pushed, to push mesh to the target terminal of user to be pushed
Mark information.
When the quantity of processed user is zero, the first probability threshold value can be initial threshold.Pushed away using initial threshold
Send judgement and respective handling.After handling a part of user, in order to avoid the number of users that finally pushes with it is default
There is larger difference in number of users, according to the destination number that user has been pushed in processed user and can push user's
Desired amt adjusts probability threshold value, and the purpose of adjustment is in order that quantity is actually pushed in all users pushes quantity phase with predetermined
Same or gap is within the specified range.
Wherein, processed user is to have carried out pushing the user judged with respective handling, and processed user is included
Push user and do not push user;Untreated user refers to not carry out pushing the user judged with respective handling.Push and judge
It can include step 202 and step 203 with respective handling.The destination number for having pushed user is actual in processed user pushed away
Send the quantity of the user of target information.The desired amt for having pushed user is to expect in processed user to be pushed target information
The quantity of user.Destination number and desired amt are compared, are adjustable probability threshold value according to comparative result.
Based on this, the disclosure can will push the destination number of user as the influence factor of the first probability threshold value.Tool
Body, it can be compared according to the destination number for having pushed user and the desired amt for having pushed user, according to comparative result
Obtain the first probability threshold value.
It is determined that after the first probability threshold value, for each untreated user, the untreated user can be obtained to target
The click probability of information;If clicking on probability is more than or equal to first probability threshold value, by the terminal of the untreated user
It is defined as the corresponding target terminal of user to be pushed, and corresponding push operation is performed for target terminal.
On clicking on probability, it is the probability that user checks the target information to click on probability, and user is often through click target
The mode of information checks target information, therefore is properly termed as checking probability or clicks on probability.Different user is to the emerging of target information
Interesting degree is different, therefore it is also different to click on probability.For example, information characteristics and user characteristics can be based on, study obtains logic
Regression model, click probability of the user to target information is calculated using Logic Regression Models.It is understood that for calculating
User can also use the other modes in correlation technique, not limit herein to the click probability of target information.
After click probability of the untreated user to target information is obtained, probability can be will click on and the first probability threshold value enters
Row compares, and judges whether the untreated user is user to be pushed according to comparative result, so as to obtain user's correspondence to be pushed
Target terminal.Wherein, target terminal can be terminal identification information.If specifically, it is general more than or equal to first to click on probability
During rate threshold value, the terminal of the user is defined as target terminal, and target information is pushed to target terminal;If clicking on probability to be less than
During the first probability threshold value, the terminal for judging the user is not target terminal, then is handled without push.Can be by the present embodiment
The processing procedure, which is referred to as pushing, to be judged and respective handling.It can be carried out pushing for each untreated user and judge and corresponding position
Reason.
Judge and corresponding it is understood that to each untreated user can push using the first probability threshold value
Processing, can also be after carrying out pushing judgement and respective handling, again according to the mesh for having pushed user to the untreated user in part
Quantity acquisition probability threshold value is marked, and push is carried out to remaining untreated user using the probability threshold value newly obtained and is judged and corresponding
Processing.
As seen from the above-described embodiment, the disclosure obtains the first probability threshold value according to the destination number for having pushed user and performed
Corresponding push operation, without just determining probability threshold value after the completion of the click probability calculation of all users and performing corresponding push
Operation, so as to improve the real-time of pushed information.Simultaneously as click probability point of the user to target information in all users
Cloth is uneven, therefore updates the first probability threshold value by the destination number for having pushed user, can improve the accurate of probability threshold value
Property, make actually to push quantity in all users and be more nearly with the predetermined quantity that pushes.
In an optional implementation, all users that will can be handled are grouped, in one group
Each user, which according to probability threshold value push, to be judged, with after respective handling, to be performed to general according to the destination number for having pushed user
Rate threshold value updates operation (step 201), then to all users in next group according to the probability threshold value progress push judgement after renewal
With respective handling (step 202 and step 203).
For example, all users can be divided into K parts, also known as K groups.Dividing mode can be that decile is divided, and also may be used
Divided with not decile.For example, being grouped by 20%, 20%, 30%, 30%, set with specific reference to demand.
In one example, it can be obtained from the K parts and be not carried out the of push operation according to the order by 1 to K
K parts, k ∈ [1, K].User's correspondence to be pushed described in being determined according to first probability threshold value from the kth certain customers
Target terminal.
Wherein, the group/part for pushing and judging with the group/part, referred to as executed push operation of respective handling is performed, is not held
Row pushes the group/part judged with respective handling, is referred to as the group/part for being not carried out push operation.The embodiment is by all users
K parts are divided into, step 201 can be performed to step 203 to the user of every part successively, until all users have been handled
Into.
It can be seen that, all users are divided into K groups in advance, the first probability threshold value is obtained according to the destination number for having pushed user,
Current group user is performed to push using the first probability threshold value and judged and respective handling.Handled to currently organizing all users
Afterwards, then can be according to the quantity update probability threshold value for having pushed user, under because the quantity for having pushed user changes
One group information pushes the probability threshold value in judging, so as to realize dynamic update probability threshold value, improves the accuracy of probability threshold value, enters
And make actually to push quantity in all users and be more nearly with the predetermined quantity that pushes.
In one example, the destination number for having pushed user can push number of users in all users.If
Current group is kth group, then the destination number for having pushed user is actually to push number of users in preceding k-1 groups, has pushed user's
Desired amt is expectation push number of users in preceding k-1 groups.
The embodiment will push number of users as the influence factor for determining the first probability threshold value in all users, can be with
Improve the accuracy for determining the first probability threshold value.
In another example, pushed user destination number can be the part of kth -1 in pushed number of users.Such as
The current group of fruit is kth group, then the destination number for having pushed user is actually to push number of users in the group of kth -1, has pushed user
Desired amt be to expect in the group of kth -1 to push number of users.
The embodiment will push number of users as the influence factor for determining the first probability threshold value in upper one group of user, real
The renewal of existing probability threshold value.
If k=1, when the quantity for representing processed user is zero, the first probability threshold value can be initial threshold.
In one example, initial threshold can be preassigned probability threshold value, and different types of target information may be used also
To set different click probability threshold values, specifically can flexibly it set.
It can be seen that, initial threshold is obtained by default mode, the efficiency for obtaining initial threshold can be improved.
, can be based on sample user group in order to improve the correlation of initial threshold and target information in another example
Middle user is to the click determine the probability initial threshold of target information, if specifically, k=1, the basis has pushed the mesh of user
Mark quantity and obtain the first probability threshold value, including:
The user of predetermined ratio is randomly selected from all users, sample user group is obtained.
Calculate each user in the sample user group and click on probability for the first of target information.
Determine that first is general according to push quantity is made a reservation in the described first click probability, the predetermined ratio and all users
Rate threshold value.
Wherein, predetermined ratio can be the oversampling ratio pre-set, can be flexible according to the total number of users of all users
Configuration.In one example, predetermined ratio can be 1%, if total number of users is very big, predetermined ratio can be 0.1%.On
User is extracted, can extract user's mark, for example, extracting the user account of preset ratio from all user names.
It can be seen that, the user of predetermined ratio is randomly selected from all users, only the first of calculating section user clicks on probability,
And click on the probability threshold value of determine the probability first using first, it is possible to reduce amount of calculation.Further, since the first probability threshold value and target
The correlation of information is higher, then the first probability threshold value that this mode is obtained is compared to the threshold value directly specified, and accuracy is higher.
Determine that first is general according to push quantity is made a reservation in the described first click probability, the predetermined ratio and all users
Rate threshold value, in order to when click on probability filtering using the first probability threshold value, all users can be made to the target
In the click probability of information, more than or equal to the quantity of the click probability of first probability threshold value, with predetermined push quantity phase
Same or gap is within the specified range.Specified range is the error range of permission set in advance.
One of which determination mode is exemplified below to illustrate.
It is described to determine the according to making a reservation for push quantity in the described first click probability, the predetermined ratio and all users
One probability threshold value, including:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained.
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets.
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
Wherein, it is to expect to push target information the need for filtering out from all users to make a reservation for push quantity in all users
Total number of users, predetermined push quantity is multiplied with predetermined ratio, you can determine to expect the number of users of push in sample user group
Amount, then can be by probability due to each user in sample user group being ranked up for the first click probability of target information
It is the first probability threshold value that the first of ranking (N*R), which clicks on determine the probability, in set.
It can be seen that, by the way that the first click probability is ranked up by order from big to small, Making by Probability Sets is obtained, and by probability
It is the first probability threshold value that the first of ranking (N*R), which clicks on determine the probability, in set, and the first probability threshold is obtained so as to quick
Value.
If k ∈ (1, K], represent that the quantity of processed user is not zero.Use can have been pushed according in processed user
The destination number at family and the desired amt update probability threshold value for having pushed user.
In one example, the basis has pushed destination number the first probability threshold value of acquisition of user, including:
Obtain the destination number for having pushed user.
When the destination number is more than N0, the second probability threshold value of history is reconfigured, the first probability threshold value, described is obtained
One probability threshold value is more than second probability threshold value.
When the destination number is less than N0, the second probability threshold value of history is reconfigured, the first probability threshold value, described is obtained
One probability threshold value is less than second probability threshold value.
Wherein, the N0To have pushed the desired amt of user.First probability threshold value and the second probability threshold value are probability
Threshold value, in order to distinguish the probability threshold value before and after updating, is referred to as the second probability threshold value, after renewal by the probability threshold value before renewal
Probability threshold value is referred to as the first probability threshold value.
It can be seen that, in this embodiment, if destination number is more than desired amt, expression has been filtered out many using probability threshold value
Remaining user, therefore probability threshold value is tuned up.For example, the second probability threshold value in history is updated to the first probability threshold value, and
First probability threshold value is more than second probability threshold value.If destination number is less than desired amt, represent to utilize probability threshold
Number of users that value is filtered out not enough, therefore will click on probability threshold value and turn down.For example, the second probability threshold value in history is updated
For the first probability threshold value, and first probability threshold value is less than second probability threshold value.If destination number, which is equal to, expects number
Amount, represents suitable using the number of users that probability threshold value filters out is clicked on, then need not adjust probability threshold value, the first probability threshold value with
Second probability threshold value is identical.
Wherein, the adjustment amplitude of probability threshold value, can be determined by default Dynamic gene.The Dynamic gene that tunes up and turn down
Dynamic gene can be with identical, can also be different, illustrated below so that two kinds of Dynamic genes are identical as an example.
In one example, when the destination number is more than N0, using formula h1=h*a reconfigures the second probability of history
Threshold value;When the destination number is less than N0, using formula h1=ha reconfigures the second probability threshold value of history;Wherein, h1Represent institute
The first probability threshold value is stated, h represents second probability threshold value, and a represents default Dynamic gene, a > 1.
Default Dynamic gene is controlling elements of regulating the speed, and the factor is bigger, and the speed of adjustment is faster, but also more unstable,
It therefore, it can configure according to demand.In one example, default Dynamic gene can be fixed value set in advance.Such as a=
1.1.In another example, default Dynamic gene can according to destination number and the gap of desired amt flexible configuration.Example
Such as, gap is bigger, and default Dynamic gene is bigger, and gap is smaller, presets Dynamic gene smaller.
It can be seen that, by the way that probability threshold value is multiplied with the default Dynamic gene more than 1, so as to increase probability threshold value, by inciting somebody to action
Probability threshold value is divided by with the default Dynamic gene more than 1, so as to reduce probability threshold value, realizes quick update probability threshold value.
In another example, the second probability threshold value of history can also be reconfigured using other modes, for example, when described
Destination number is more than N0, using formula h1=h+a reconfigures the second probability threshold value of history, when the destination number is less than N0, adopt
Use formula h1=h-a reconfigures the second probability threshold value of history.Mode is reconfigured for other, this is no longer going to repeat them.
Desired amt on having pushed user, desired amt is to expect in processed user to be pushed the use of target information
The quantity at family.It is exemplified below the method for several determination desired amts.
In one example, if the destination number is has pushed number of users in all users, desired amt is
Expect to push number of users in all users.
Now,N represents to make a reservation for push quantity in all users.
It can be seen that, the embodiment directly will make a reservation for push quantity in all users and processed user's ratio is multiplied, soon
Speed obtains desired amt.
Particularly, if all users are divided into K parts, using formulaObtain desired amt,
The accuracy for determining desired amt can be improved, and then improves the accuracy for determining the first probability threshold value.
In another example, if the destination number be all users in pushed number of users, desired amt
To expect to push number of users in user used.
Now,M represents all numbers of users, MiI-th section number of users is represented, N represents all
Make a reservation for push quantity in user.
Due to there may be situation of the K parts for not decile, then the ratio per group user number with total number of users is first determined
ValueThe desired amt of this group of user is determined according to ratioThe desired amt of processed each group user is added againIt can obtain the desired amt that user has been pushed in processed user.
It can be seen that, the desired amt obtained using above-mentioned formula, accuracy is high, and then improves the standard for determining the first probability threshold value
True property.
In one example, if the destination number is has pushed number of users in the part of kth -1, desired amt is
Expect to push number of users in the part of kth -1.
Now,N represents to make a reservation for push quantity in all users.
The present embodiment will directly make a reservation for push quantity and be divided by with K in all users, quickly obtain desired amt.Particularly,
If all users are divided into K parts, using formulaDesired amt is obtained, determination desired amt can be improved
Accuracy, and then improve determine the first probability threshold value accuracy.
In another example, if the destination number be the part of kth -1 in pushed number of users, desired amt
To expect to push number of users in the part of kth -1.
Now,M represents all numbers of users, Mk-1The certain customers' quantity of kth -1 is represented, N represents institute
Have and make a reservation for push quantity in user.
All users are divided into the situation of K parts due to there may be using not decile dividing mode, then first determine the
K-1 certain customers quantity and the ratio of total number of usersThe desired amt of this group of user is determined according to ratio
It can obtain N0。
It can be seen that, the desired amt obtained using above-mentioned formula, accuracy is high, and then improves the standard for determining the first probability threshold value
True property.
After the first probability threshold value is obtained, for each user in kth certain customers, the user is obtained to target information
Second click on probability;The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, it is determined that
For the corresponding target terminal of user to be pushed, and target information is pushed to the target terminal.
Various technical characteristics in embodiment of above can be arbitrarily combined, as long as the combination between feature is not present
Conflict or contradiction, but as space is limited, are not described one by one, therefore the various technical characteristics in above-mentioned embodiment is any
It is combined the scope for falling within this disclosure.
The disclosure is enumerated one of combination and illustrated.As shown in figure 3, Fig. 3 is that the disclosure is exemplary according to one
Implement the flow chart of another information-pushing method exemplified, methods described includes:
In step 301, the user of predetermined ratio is randomly selected from all users, sample user group is obtained.
In step 302, calculate each user in the sample user group and click on probability for the first of target information.
In step 303, clicked on according to described first and make a reservation for push number in probability, the predetermined ratio and all users
Amount determines the first probability threshold value.
In step 304, all users are carried out being divided into K groups.
In step 305, according to the order by 1 to K, the kth group for being not carried out push operation, root are obtained from the K groups
Determine from the kth group user the corresponding target terminal of the user to be pushed according to first probability threshold value, and to target
Terminal pushes target information.
Wherein, during k=1, the first probability threshold value used can determined by step 303 threshold value, k ∈ (1, K] when,
The first probability threshold value used has pushed destination number and the phase of user in the preceding k-1 groups user based on executed push operation
Quantity is hoped to obtain.
, can described in being determined according to the first probability threshold value from kth group user when push user corresponding target terminal
For each user in kth group user, to obtain the user and click on probability to the second of target information, if the second click probability is big
In or equal to the first probability threshold value, then the terminal of the user is defined as to the target terminal of user to be pushed, and it is whole to the target
End pushes target information.Above-mentioned push judgement and respective handling are performed to each user of kth group successively, until the institute of kth group
There is user's processing completion.It can be seen that, for each user, click on determine the probability and complete, corresponding push processing is also completed.
Within step 306, if k ≠ K, all users according to the first probability threshold value push and judged simultaneously in kth group
Carry out after respective handling, the destination number and desired amt of user will have been pushed in the preceding k-1 groups user of executed push operation
It is compared, first probability threshold value is updated according to comparative result, k=k+1, and return to step 305 is made, to utilize renewal
Probability threshold value afterwards carries out pushing judgement and respective handling to user in next group.
In step 307, if k=K, push processing terminates.
As seen from the above-described embodiment, the disclosure is based on sample user group and estimates out probability threshold value, and in every group of user's processing
After complete, probability threshold value is updated, it is to avoid the number of users finally pushed has lacking for larger difference with predesignated subscriber's quantity
Fall into.Meanwhile, carry out pushing judgement for each user in every group and corresponding push is handled, improve push real-time.
Understand for convenience, illustrated below with a concrete application example.Assuming that total number of users of all users
It is 1% to make a reservation for push quantity N, predetermined ratio in M, all users.The user of random selection 1% from all users, obtains
The sample user group of M/100 numbers, calculates the click probability of each user in sample user group to target information, by sample user
The click probability of each user is ranked up by size in group, Making by Probability Sets is obtained, by the point of No. N/100 of ranking in Making by Probability Sets
Determine the probability is hit for probability threshold value h.
All users are divided into K deciles.For first group of user, determine that click of the active user to target information is general
Rate;It will click on probability and be compared with clicking on probability threshold value h, the user to the active user is judged whether according to comparative result
Terminal pushes the target information, and performs corresponding push processing.When all users are carried out according to probability threshold value h in first group
Push and judge and carry out after respective handling, the number of users n that will actually be pushed in first group1WithIt is compared, ifProbability threshold value is then updated to h1=h*1.1, ifProbability threshold value is then updated to h1=h/1.1, and
To all users in second group according to probability threshold value h1Carry out pushing judgement and carry out respective handling, by that analogy.When 1 to K-1
After the completion of group user calculates, the actual number of users pushed is n1~K-1If,Then by probability threshold value more
It is newly hK-1=hK-2* 1.1, ifProbability threshold value is then updated to hK-1=hK-2/ 1.1, and to K groups
Interior all users are according to probability threshold value hK-1Carry out pushing judgement and respective handling.
Embodiment with aforementioned information method for pushing is corresponding, and the disclosure additionally provides information push-delivery apparatus and its applied
Service end embodiment.
As shown in figure 4, Fig. 4 is a kind of block diagram of information push-delivery apparatus of the disclosure according to an exemplary embodiment,
Described device includes:Threshold value acquisition module 41, terminal deciding module 42 and push operation module 43.
Wherein, threshold value acquisition module 41, is configured as obtaining the first probability threshold value according to the destination number for having pushed user.
Terminal deciding module 42, is configured as determining that the corresponding target of user to be pushed is whole according to first probability threshold value
End.
Push operation module 43, is configured as performing push operation for the target terminal.
As seen from the above-described embodiment, obtain the first probability threshold value according to the destination number for having pushed user and perform and accordingly push away
Operation is sent, without just determining probability threshold value after the completion of the click probability calculation of all users and performing corresponding push operation,
So as to improve the real-time of pushed information.Simultaneously as in all users user to the click probability distribution of target information not
Uniformly, therefore by the destination number for having pushed user the first probability threshold value is updated, the accuracy of probability threshold value can be improved, made
Quantity is actually pushed in all users to be more nearly with the predetermined quantity that pushes.
As shown in figure 5, Fig. 5 is the frame of another information push-delivery apparatus of the disclosure according to an exemplary embodiment
Figure, the embodiment is on the basis of foregoing embodiment illustrated in fig. 4, and all users are divided into K parts;The terminal deciding module
42, including:Acquisition submodule 421 and terminal determination sub-module 422.
Wherein, acquisition submodule 421, are configured as according to the order by 1 to K, are obtained from the K parts and are not carried out pushing away
Send the kth part of operation, k ∈ [1, K].
Terminal determination sub-module 422, is configured as being determined from the kth certain customers according to first probability threshold value
The corresponding target terminal of the user to be pushed.
As seen from the above-described embodiment, all users are divided into K parts, the is obtained according to the destination number for having pushed user
One probability threshold value, is performed to current group user using the first probability threshold value and pushes judgement and respective handling.All to current group
After user is handled, because the quantity for having pushed user changes, then it can update general according to the quantity for having pushed user
Rate threshold value, the probability threshold value in judging is pushed as next group information, so as to realize dynamic update probability threshold value, improves probability threshold
The accuracy of value, and then make actually to push quantity in all users and be more nearly with the predetermined quantity that pushes.
In an optional implementation, the destination number is to have pushed number of users in all users, or described
Destination number be the part of kth -1 in pushed number of users.
As seen from the above-described embodiment, number of users will have been pushed in all users as the influence for determining the first probability threshold value
Factor, can improve the accuracy for determining the first probability threshold value.
As shown in fig. 6, Fig. 6 is the frame of another information push-delivery apparatus of the disclosure according to an exemplary embodiment
Figure, the embodiment on the basis of foregoing embodiment illustrated in fig. 4, k ∈ (1, K], the threshold value acquisition module 41 includes:Quantity is obtained
Submodule 411 and threshold value is taken to reconfigure submodule 412.
Wherein, quantity acquisition submodule 411, is configured as obtaining the destination number for having pushed user.
Threshold value reconfigures submodule 412, is configured to when the destination number is more than N0, reconfigure the second probability threshold of history
Value, obtains the first probability threshold value, and first probability threshold value is more than second probability threshold value;When the destination number is less than
N0, the second probability threshold value of history is reconfigured, the first probability threshold value is obtained, first probability threshold value is less than second probability
Threshold value;Wherein, the N0To have pushed the desired amt of user.
As seen from the above-described embodiment, the target of user will have been pushed in processed user, the expectation number with having pushed user
Amount is compared, according to comparative result update probability threshold value, so that the probability threshold value after being updated, realizes quick update probability
Threshold value.
In an optional implementation, if the destination number is has pushed number of users in all users,Or
Wherein, M represents all numbers of users, MiI-th section number of users is represented, N represents to make a reservation for push in all users
Quantity.
In an optional implementation, if the destination number is has pushed number of users in the part of kth -1,Or
Wherein, M represents all numbers of users, Mk-1The certain customers' quantity of kth -1 is represented, N represents to make a reservation for push away in all users
Send quantity.
In an optional implementation, the threshold value reconfigures submodule 412, and concrete configuration is:
When the destination number is more than N0, using formula h1=h*a reconfigures the second probability threshold value of history.
When the destination number is less than N0, using formula h1=h/a reconfigures the second probability threshold value of history.
Wherein, h1First probability threshold value is represented, h represents second probability threshold value, and a represents default Dynamic gene, a
> 1.
As seen from the above-described embodiment, by the way that probability threshold value is multiplied with the default Dynamic gene more than 1, so as to increase probability
Threshold value, by the way that probability threshold value is divided by with the default Dynamic gene more than 1, so as to reduce probability threshold value, realizes quick update generally
Rate threshold value.
As shown in fig. 7, Fig. 7 is the frame of another information push-delivery apparatus of the disclosure according to an exemplary embodiment
Figure, the embodiment is on the basis of foregoing embodiment illustrated in fig. 4, if k=1, and the threshold value acquisition module 41 includes:Customer group
Sampling submodule 413, probability calculation submodule 414 and threshold value determination sub-module 415.
Wherein, customer group sampling submodule 413, is configured as randomly selecting the user of predetermined ratio from all users,
Obtain sample user group.
Probability calculation submodule 414, is configured as calculating each user in the sample user group and is directed to target information
First clicks on probability.
Threshold value determination sub-module 415, is configured as useful according to the described first click probability, the predetermined ratio and institute
Make a reservation for push quantity in family and determine the first probability threshold value.
As seen from the above-described embodiment, the user of predetermined ratio is randomly selected from all users, only calculating section user
First clicks on probability, and clicks on the probability threshold value of determine the probability first using first, it is possible to reduce amount of calculation.Further, since first
Probability threshold value and the correlation of target information are higher, then the first probability threshold value that this mode is obtained is compared to the threshold directly specified
Value, accuracy is higher.
In an optional implementation, the threshold value determination sub-module 415, concrete configuration is:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained.
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets.
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
In an optional implementation, the terminal determination sub-module 422 is specifically configured to:
For each user in kth certain customers, obtain the user and click on probability to the second of target information.
The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, is defined as waiting to push
The corresponding target terminal of user.
Accordingly, the disclosure also provides a kind of information push-delivery apparatus, and described device includes processor;For storage processing
The memory of device executable instruction;Wherein, the processor is configured as:
Destination number according to user has been pushed obtains the first probability threshold value.
The corresponding target terminal of user to be pushed is determined according to first probability threshold value.
Push operation is performed for the target terminal.
The function of modules and the specific details of implementation process of effect are shown in correspondence step in the above method in said apparatus
Implementation process, will not be repeated here.
For device embodiment, because it corresponds essentially to embodiment of the method, so related part is real referring to method
Apply the part explanation of example.Device embodiment described above is only schematical, wherein described be used as separating component
The module of explanation can be or may not be physically separate, and the part shown as module can be or can also
It is not physical module, you can with positioned at a place, or can also be distributed on multiple mixed-media network modules mixed-medias.Can be according to reality
Selection some or all of module therein is needed to realize the purpose of disclosure scheme.Those of ordinary skill in the art are not paying
In the case of going out creative work, you can to understand and implement.
As shown in figure 8, Fig. 8 is a kind of frame of device 800 pushed for information according to an exemplary embodiment
Figure.
For example, device 800 may be provided in server apparatus.Reference picture 8, system 800 includes processing assembly 822, its
Further comprise one or more processors, and as the memory resource representated by memory 832, can be by handling for storing
The instruction of the execution of part 822, such as application program.The application program stored in memory 832 can include one or one
Above each corresponds to the module of one group of instruction.In addition, processing assembly 822 is configured as execute instruction, it is above-mentioned to perform
Information-pushing method.
System 800 can also include the power management that a power supply module 826 is configured as execution system 800, and one has
Line or radio network interface 850 are configured as system 800 being connected to network, and input and output (I/O) interface 858.System
System 800 can be operated based on the operating system for being stored in memory 832.
Wherein, when the instruction in the memory 832 is performed by the processing assembly 822 so that system 800 can be held
A kind of information-pushing method of row, including:
Destination number according to user has been pushed obtains the first probability threshold value.
The corresponding target terminal of user to be pushed is determined according to first probability threshold value.
Push operation is performed for the target terminal.
Those skilled in the art will readily occur to its of the disclosure after considering specification and putting into practice invention disclosed herein
Its embodiment.The disclosure is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or
Person's adaptations follow the general principle of the disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.Description and embodiments are considered only as exemplary, and the true scope of the disclosure and spirit are by following
Claim is pointed out.
It should be appreciated that the precision architecture that the disclosure is not limited to be described above and is shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present disclosure is only limited by appended claim.
The preferred embodiment of the disclosure is the foregoing is only, not to limit the disclosure, all essences in the disclosure
God is with principle, and any modification, equivalent substitution and improvements done etc. should be included within the scope of disclosure protection.
Claims (19)
1. a kind of information-pushing method, it is characterised in that methods described includes:
Destination number according to user has been pushed obtains the first probability threshold value;
The corresponding target terminal of user to be pushed is determined according to first probability threshold value;
Push operation is performed for the target terminal.
2. according to the method described in claim 1, it is characterised in that all users are divided into K parts;
It is described that the corresponding target terminal of user to be pushed is determined according to first probability threshold value, including:
According to the order by 1 to K, the kth part for being not carried out push operation, k ∈ [1, K] are obtained from the K parts;
The corresponding target terminal of user to be pushed described in being determined according to first probability threshold value from the kth certain customers.
3. method according to claim 2, it is characterised in that the destination number is to have pushed number of users in all users
Amount, or the destination number be the part of kth -1 in pushed number of users.
4. method according to claim 2, it is characterised in that and k ∈ (1, K], the basis has pushed the number of targets of user
Amount obtains the first probability threshold value, including:
Obtain the destination number for having pushed user;
When the destination number is more than N0, the second probability threshold value of history is reconfigured, the first probability threshold value is obtained, described first is general
Rate threshold value is more than second probability threshold value;
When the destination number is less than N0, the second probability threshold value of history is reconfigured, the first probability threshold value is obtained, described first is general
Rate threshold value is less than second probability threshold value;
Wherein, the N0To have pushed the desired amt of user.
5. method according to claim 4, it is characterised in that if the destination number is to have pushed use in all users
Amount amount,Or
If the destination number is has pushed number of users in the part of kth -1,Or
Wherein, M represents all numbers of users, MiRepresent i-th section number of users, Mk-1Represent the certain customers' quantity of kth -1, N tables
Show and make a reservation for push quantity in all users.
6. method according to claim 4, it is characterised in that
When the destination number is more than N0, using formula h1=h*a reconfigures the second probability threshold value of history;
When the destination number is less than N0, using formula h1=h/a reconfigures the second probability threshold value of history;
Wherein, h1First probability threshold value is represented, h represents second probability threshold value, and a represents default Dynamic gene, a > 1.
7. method according to claim 2, it is characterised in that if k=1, the basis has pushed the number of targets of user
Amount obtains the first probability threshold value, including:
The user of predetermined ratio is randomly selected from all users, sample user group is obtained;
Calculate each user in the sample user group and click on probability for the first of target information;
The first probability threshold is determined according to push quantity is made a reservation in the described first click probability, the predetermined ratio and all users
Value.
8. method according to claim 7, it is characterised in that described to click on probability, the predetermined ratio according to described first
Example and all users in make a reservation for push quantity determine the first probability threshold value, including:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained;
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets;
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
9. method according to claim 2, it is characterised in that it is described according to first probability threshold value from the kth portion
Divide in user and determine the corresponding target terminal of the user to be pushed, including:
For each user in kth certain customers, obtain the user and click on probability to the second of target information;
The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, is defined as user to be pushed
Corresponding target terminal.
10. a kind of information push-delivery apparatus, it is characterised in that described device includes:
Threshold value acquisition module, is configured as obtaining the first probability threshold value according to the destination number for having pushed user;
Terminal deciding module, is configured as determining the corresponding target terminal of user to be pushed according to first probability threshold value;
Push operation module, is configured as performing push operation for the target terminal.
11. device according to claim 10, it is characterised in that all users are divided into K parts;The terminal is determined
Module, including:
Acquisition submodule, is configured as according to the order by 1 to K, and the kth for being not carried out push operation is obtained from the K parts
Part, k ∈ [1, K];
Terminal determination sub-module, be configured as being determined from the kth certain customers according to first probability threshold value described in treat
Push the corresponding target terminal of user.
12. device according to claim 11, it is characterised in that the destination number is to have pushed user in all users
Quantity, or the destination number be the part of kth -1 in pushed number of users.
13. device according to claim 11, it is characterised in that and k ∈ (1, K], the threshold value acquisition module includes:
Quantity acquisition submodule, is configured as obtaining the destination number for having pushed user;
Threshold value reconfigures submodule, is configured to when the destination number is more than N0, the second probability threshold value of history is reconfigured, is obtained
First probability threshold value, first probability threshold value is more than second probability threshold value;When the destination number is less than N0, reconfigure
Second probability threshold value of history, obtains the first probability threshold value, and first probability threshold value is less than second probability threshold value;Its
In, the N0To have pushed the desired amt of user.
14. device according to claim 13, it is characterised in that if the destination number is to have pushed in all users
Number of users,Or
If the destination number is has pushed number of users in the part of kth -1,Or
Wherein, M represents all numbers of users, MiRepresent i-th section number of users, Mk-1Represent the certain customers' quantity of kth -1, N tables
Show and make a reservation for push quantity in all users.
15. device according to claim 13, it is characterised in that the threshold value reconfigures submodule, concrete configuration is:
When the destination number is more than N0, using formula h1=h*a reconfigures the second probability threshold value of history;
When the destination number is less than N0, using formula h1=h/a reconfigures the second probability threshold value of history;
Wherein, h1First probability threshold value is represented, h represents second probability threshold value, and a represents default Dynamic gene, a > 1.
16. device according to claim 11, it is characterised in that if k=1, the threshold value acquisition module includes:
Customer group sampling submodule, is configured as randomly selecting the user of predetermined ratio from all users, obtains sample user
Group;
Probability calculation submodule, is configured as calculating each user in the sample user group and is clicked on for the first of target information
Probability;
Threshold value determination sub-module, is configured as clicking in probability, the predetermined ratio and all users in advance according to described first
Surely push quantity and determine the first probability threshold value.
17. device according to claim 16, it is characterised in that the threshold value determination sub-module, concrete configuration is:
Described first click probability is ranked up by order from big to small, Making by Probability Sets is obtained;
It is the first probability threshold value to click on determine the probability by first of ranking (N*R) in the Making by Probability Sets;
Wherein, N represents to make a reservation for push quantity in all users, and R represents the predetermined ratio.
18. device according to claim 11, it is characterised in that the terminal determination sub-module, is specifically configured to:
For each user in kth certain customers, obtain the user and click on probability to the second of target information;
The terminal that probability is more than or equal to the user of first probability threshold value is clicked on by described second, is defined as user to be pushed
Corresponding target terminal.
19. a kind of information push-delivery apparatus, it is characterised in that including:
Processor;
Memory for storing processor-executable instruction;
Wherein, the processor is configured as:
Destination number according to user has been pushed obtains the first probability threshold value;
The corresponding target terminal of user to be pushed is determined according to first probability threshold value;
Push operation is performed for the target terminal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710125117.6A CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710125117.6A CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106982250A true CN106982250A (en) | 2017-07-25 |
CN106982250B CN106982250B (en) | 2020-06-02 |
Family
ID=59338729
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710125117.6A Active CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106982250B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171267A (en) * | 2017-12-28 | 2018-06-15 | 广州优视网络科技有限公司 | User group partitioning method and device, information push method and device |
CN111131359A (en) * | 2018-10-31 | 2020-05-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
CN111273939A (en) * | 2020-01-20 | 2020-06-12 | Oppo(重庆)智能科技有限公司 | Information processing method, information processing device and terminal equipment |
CN112328947A (en) * | 2020-12-31 | 2021-02-05 | 腾讯科技(深圳)有限公司 | Distribution method and device of to-be-recommended media information and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100332330A1 (en) * | 2009-06-30 | 2010-12-30 | Google Inc. | Propagating promotional information on a social network |
CN105335491A (en) * | 2015-10-20 | 2016-02-17 | 杭州东信北邮信息技术有限公司 | Method and system for recommending books to users on basis of clicking behavior of users |
CN105915438A (en) * | 2016-04-15 | 2016-08-31 | 北京奇虎科技有限公司 | Message pushing method, apparatus, and system |
-
2017
- 2017-03-03 CN CN201710125117.6A patent/CN106982250B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100332330A1 (en) * | 2009-06-30 | 2010-12-30 | Google Inc. | Propagating promotional information on a social network |
CN105335491A (en) * | 2015-10-20 | 2016-02-17 | 杭州东信北邮信息技术有限公司 | Method and system for recommending books to users on basis of clicking behavior of users |
CN105915438A (en) * | 2016-04-15 | 2016-08-31 | 北京奇虎科技有限公司 | Message pushing method, apparatus, and system |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171267A (en) * | 2017-12-28 | 2018-06-15 | 广州优视网络科技有限公司 | User group partitioning method and device, information push method and device |
CN108171267B (en) * | 2017-12-28 | 2022-03-22 | 阿里巴巴(中国)有限公司 | User group division method and device and message pushing method and device |
CN111131359A (en) * | 2018-10-31 | 2020-05-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
CN111131359B (en) * | 2018-10-31 | 2022-06-07 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
CN111273939A (en) * | 2020-01-20 | 2020-06-12 | Oppo(重庆)智能科技有限公司 | Information processing method, information processing device and terminal equipment |
CN111273939B (en) * | 2020-01-20 | 2023-11-10 | Oppo(重庆)智能科技有限公司 | Information processing method, information processing device and terminal equipment |
CN112328947A (en) * | 2020-12-31 | 2021-02-05 | 腾讯科技(深圳)有限公司 | Distribution method and device of to-be-recommended media information and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106982250B (en) | 2020-06-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106982250A (en) | Information-pushing method and device | |
CN109857935B (en) | Information recommendation method and device | |
CN110413867B (en) | Method and system for content recommendation | |
WO2019019649A1 (en) | Method and apparatus for generating investment portfolio product, storage medium and computer device | |
CN106227881B (en) | Information processing method and server | |
CN104765793A (en) | Software recommending method and server | |
CN104408640A (en) | Application software recommending method and apparatus | |
CN104391996A (en) | Method for collecting and analyzing user interaction behavior of mobile terminal | |
CN108875035B (en) | Data storage method of distributed file system and related equipment | |
CN108667877B (en) | Method and device for determining recommendation information, computer equipment and storage medium | |
CN110969240A (en) | Pruning method, device, equipment and medium for deep convolutional neural network | |
CN110457704B (en) | Target field determination method and device, storage medium and electronic device | |
CN110737458A (en) | code updating method and related device | |
CN112364185A (en) | Method and device for determining characteristics of multimedia resource, electronic equipment and storage medium | |
CN104010041A (en) | Control method and device for recommending APP in application program | |
CN110909072B (en) | Data table establishment method, device and equipment | |
CN110035126A (en) | A kind of document handling method, calculates equipment and storage medium at device | |
CN108090782B (en) | Online game recommendation method and server | |
CN107784363B (en) | Data processing method, device and system | |
CN105468603B (en) | Data selecting method and device | |
CN103810157A (en) | Method and device for achieving input method | |
CN105069064B (en) | Acquisition methods and device, the method for pushing and device of vocabulary | |
CN112163712A (en) | Catering processing method and device, storage medium and catering system | |
CN109241348B (en) | Video recommendation method and system | |
CN108989181B (en) | Method, device, equipment and server for establishing instant messaging |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |