CN109379410B - Information-pushing method, device, server and storage medium - Google Patents
Information-pushing method, device, server and storage medium Download PDFInfo
- Publication number
- CN109379410B CN109379410B CN201811117810.XA CN201811117810A CN109379410B CN 109379410 B CN109379410 B CN 109379410B CN 201811117810 A CN201811117810 A CN 201811117810A CN 109379410 B CN109379410 B CN 109379410B
- Authority
- CN
- China
- Prior art keywords
- information
- push
- pushed
- probability
- indicate
- 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.)
- Active
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
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The application is about a kind of information-pushing method, device, server and storage medium, comprising: establishes score matrix and push amount vector according to information to be pushed, the score matrix is the matrix of component number building by information push effect quantization；Push computation model is established according to score matrix and push amount vector；Wherein, push computation model is the computation model for the parameter probability valuing that pushed information is pushed under the constraint condition for meeting score matrix and push amount vector；The push probability of the information to be pushed is calculated using push computation model；The information to be pushed is pushed according to the push probability.The information-pushing method, device, server and storage medium are by establishing push computation model, obtaining the push probability of corresponding information to be pushed and realizing the push of information to be pushed according to the push probability.Information-pushing method, device, server and storage medium can obtain the push effect of optimization under conditions of guaranteeing that push demand obtains satisfaction.
Description
Technical field
This application involves internet application field more particularly to a kind of information-pushing method, device, server and storages
Medium.
Background technique
In Internet application, information has popularization demand, it is often necessary to push information at the specified target such as client.
The owner of information or information is set to obtain a certain amount of exposure and concern.
In the related technology, main to realize that steady push is target, by taking pushing video as an example, pushing video is usually will be certain
The video of amount calculates the position of the probability and push that currently push according to the needs of push amount.Before guaranteeing push amount
It puts, realizes the steady push of video, complete the push figureofmerit of video.
But it is by adopting the above technical scheme, amount sequence according to demand to be pushed, leads to not maximize push effect.
Summary of the invention
To overcome the problems in correlation technique, the application provide a kind of information-pushing method, device, server and
Storage medium.
According to the embodiment of the present application in a first aspect, providing a kind of information-pushing method, comprising:
Score matrix and push amount vector are established according to information to be pushed, the score matrix is to push effect amount by information
Change and the matrix of component number building；
Push computation model is established according to the score matrix and push amount vector；Wherein, the push computation model is
The computation model for the parameter probability valuing that pushed information is pushed under the constraint condition for meeting the score matrix and push amount vector；
The push probability of the information to be pushed is calculated using push computation model；
The information to be pushed is pushed according to the push probability.
Optionally, the information to be pushed is pushed according to the push probability, comprising:
According to the corresponding push probability of information to be pushed, formation probability section；
Generate the random number of corresponding information to be pushed；
Judge whether random number is in the corresponding probability interval；
If random number is in the corresponding probability interval, the information to be pushed of corresponding random number is launched to accession page
In pushed, otherwise do not push.
Optionally, according to the corresponding push probability of information to be pushed, before formation probability section, further includes:
Obtain the maximum pushed information amount of accession page；
Judge whether information to be pushed quantity is less than or equal to maximum pushed information amount；
If so, execution is described according to the corresponding push probability of information to be pushed, otherwise formation probability section does not push.
Optionally, include: using the push probability that the push computation model calculates the information to be pushed
minX(-Σi,j log(1+Aij Xij))；
s.t.Σi Xij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate in i-th of accession page pushing away for j-th of information
Send probability, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
Optionally, include: using the push probability that the push computation model calculates the information to be pushed
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is Σi Xij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djExpression pushes away
The amount of sending vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate the antithesis of push probability
Variable, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij
Expression formula calculate Xij。
Optionally, it is described score matrix and push amount vector are established according to information to be pushed before, the information push side
Method further include:
Obtain the pageview of pushed information on the page；
The information to be pushed is extracted from the pushed information according to the pageview and is stored.
Optionally, the score matrix is used for characterization information clicking rate, the powder rate that rises, thumbs up rate, bean vermicelli online hours or good
The push effect of sensitivity.
Optionally, the information to be pushed includes at least one information of video, audio, image, text.
According to the second aspect of the embodiment of the present application, a kind of pushed information device is provided, comprising:
Analytical unit is configured as establishing score matrix and push amount vector according to information to be pushed, the score matrix
It is the matrix of component number building by information push effect quantization；
Modeling unit is configured as establishing push computation model according to the score matrix and push amount vector；Wherein, institute
State push computation model be under the constraint condition for meeting the score matrix and push amount vector pushed information be pushed it is general
The computation model of rate value；
Arithmetic element is configured as calculating the push probability of the information to be pushed using push computation model；
Push unit is configured as pushing the information to be pushed according to the push probability.
Optionally, the push unit includes:
Section generation unit is configured as according to the corresponding push probability of information to be pushed, formation probability section；
Random number generation unit is configurable to generate the random number of corresponding information to be pushed；
Random number judging unit is configured as judging whether random number is in the corresponding probability interval；
Execute push unit, if being configured as random number is in the corresponding probability interval, by corresponding random number to
Pushed information dispensing is pushed into accession page, is not otherwise pushed.
Optionally, the information push-delivery apparatus further include:
First acquisition unit is configured as obtaining the maximum pushed information amount of accession page；
Information content judging unit is configured as judging whether information to be pushed quantity is less than or equal to maximum pushed information
Amount；
Section generation unit is configured as if so, according to the corresponding push probability of information to be pushed, formation probability section,
Otherwise it does not push.
Optionally, include: using the push probability that the push computation model calculates the information to be pushed
minX(-Σi,j log(1+Aij Xij))；
s.t.Σi Xij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate in i-th of accession page pushing away for j-th of information
Send probability, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
Optionally, include: using the push probability that the push computation model calculates the information to be pushed
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is Σi Xij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djExpression pushes away
The amount of sending vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate the antithesis of push probability
Variable, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij
Expression formula calculate Xij。
Optionally, the information push-delivery apparatus further include:
Second acquisition unit is configured as obtaining the pageview of pushed information on the page；
Extraction unit is configured as extracting the information to be pushed from the pushed information according to the pageview and deposit
Storage.
Optionally, the score matrix is used for characterization information clicking rate, the powder rate that rises, thumbs up rate, bean vermicelli online hours or good
The push effect of sensitivity.
Optionally, the information to be pushed includes at least one information of video, audio, image, text.
According to the third aspect of the embodiment of the present application, a kind of Information Push Server is provided, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to: the step of above- mentioned information method for pushing.
According to the fourth aspect of the embodiment of the present application, a kind of non-transitorycomputer readable storage medium is provided, when described
When instruction in storage medium is executed by the processor of server, enable the server to execute such as above-mentioned pushed information side
Method.
According to the 5th of the embodiment of the present application the aspect, a kind of computer program product, including computer program code are provided,
The computer program includes program instruction, when described program instruction is computer-executed, executes the computer above-mentioned
The step of pushed information method.
The technical solution that embodiments herein provides can include the following benefits:
Score matrix and push amount vector are established according to information to be pushed；It is built according to the score matrix and push amount vector
Vertical push computation model；Wherein, the push computation model is in the constraint item for meeting the score matrix and push amount vector
The parameter probability valuing that pushed information is pushed under part；The push probability of the information to be pushed is calculated using push computation model；Root
The information to be pushed is pushed according to the push probability.The application is established push by score matrix and push amount vector and calculated
Model, and then push probability is obtained under the constraint condition for meeting the score matrix and push amount vector, utilize push probability
Pushed information can obtain preferable information push effect, i.e., under the premise of guaranteeing push amount meet demand amount, maximize real
Now push effect.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The application can be limited.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the application
Example, and together with specification it is used to explain the principle of the application.
Fig. 1 is a kind of flow chart of information-pushing method shown according to an exemplary embodiment.
Fig. 2 is a kind of embodiment process shown according to an exemplary embodiment using push probability pushed information
Figure.
Fig. 3 is a kind of reality that information to be pushed amount shown according to an exemplary embodiment combines push probability pushed information
Apply mode flow chart.
Fig. 4 is a kind of embodiment flow chart shown according to an exemplary embodiment for obtaining information to be pushed.
Fig. 5 is a kind of block diagram of information push-delivery apparatus shown according to an exemplary embodiment.
Fig. 6 is block diagram shown according to an exemplary embodiment, is to show section generation unit 641, generating random number
Unit 642, random number judging unit 643 and execution push unit 644.
Fig. 7 is the block diagram of another information push-delivery apparatus shown according to an exemplary embodiment, is to show that first obtains
Take unit 751, information content judging unit 752, section generation unit 741, random number generation unit 742, random number judging unit
743 and execute push unit 744.
Fig. 8 is the block diagram of another information push-delivery apparatus shown according to an exemplary embodiment, is to show that second obtains
Take unit 861 and extraction unit 862.
Fig. 9 is a kind of block diagram of device 900 for information push shown according to an exemplary embodiment.
Figure 10 is a kind of block diagram of device 1000 for information-pushing method shown according to an exemplary embodiment
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the application.
Fig. 1 is a kind of flow chart of information-pushing method shown according to an exemplary embodiment, as shown in Figure 1, information
Method for pushing the following steps are included:
In step s 11, score matrix and push amount vector are established according to information to be pushed, the score matrix is by believing
The matrix of breath push effect quantization and component number building.
In the present embodiment, information-pushing method can be included according to information to be pushed need push effect to be achieved and
The requirement of push amount, and quantified to form certain score, according to the score that quantization is formed, score matrix is established respectively and is pushed away
The amount of being sent to.Wherein, score matrix can reflect certain push effect, different push effects can be quantified as score, then
Score matrix is formed in conjunction with the location of information to be pushed.Push amount vector can reflect the push time of corresponding information to be pushed
Number, i.e. the push demand of information to be pushed.
Such as it is realized in webpage in the example of push, it is assumed that every access to web page webpage supports the push of single pushed information,
So in 100,000 web page access, when needing to push 100 information to be pushed, the demand of every information to be pushed is desired for
100 times.
In the process, it can also it is expected for the pointed target position of information to be pushed push in conjunction with demand, be formed
Push amount vector.Target position based on information to be pushed push is different, and the push amount vector of formation is not also identical.For example, mesh
Cursor position is set in the central area of accession page and the fringe region of accession page, it will different.
In the above process, information-pushing method can handle a plurality of pushed information simultaneously, and batch processing can also individually be located
Pushed information is managed, to promote the speed for establishing score matrix and push amount vector.
In step s 12, push computation model is established according to the score matrix and push amount vector.Wherein, described to push away
Sending computation model is under the constraint condition for meeting the score matrix and push amount vector, and the probability that pushed information is pushed takes
The computation model of value.
In the present embodiment, score matrix and the push amount vector that front is established are mainly based upon to establish push and calculate mould
Type, which needs to meet the constraint condition of the score matrix and push amount vector, for calculating pushed information
The parameter probability valuing being pushed；Under the premise of guaranteeing that push amount vector reaches requirement, the reckoning computation model can be realized phase
The push effect requirements answered, push effect requirements therein are quantified as score matrix.The reckoning computation model meets simultaneously to be pushed away
The constraint condition of the amount of sending vector sum score matrix.
In step s 13, the push probability of the information to be pushed is calculated using push computation model.
In the present embodiment, corresponding information to be pushed can be obtained according to the limitation of constraint condition in the computation model
Push probability, above-mentioned push probability reflects the probability scenarios whether information to be pushed obtains push, that is, obtain push can
It can property.
In step S14, the information to be pushed is pushed according to the push probability.
In the present embodiment, it for a web page browsing, according to the push probability that abovementioned steps calculate, is adopted according to probability
Sample pushes information to be pushed.
For example, if for certain information to be pushed push probability be 0.6, information to be pushed every 100 times push in,
Wherein pushed for 60 times.
By the information-pushing method, information to be pushed can guarantee push demand obtain satisfaction in the case where,
Realize the optimization of push effect.
Relative to directly amount meets the mode that condition successively pushes according to demand, the information-pushing method can according to point
Under the limitation of matrix number and push amount vector, the solution of optimization is obtained.Operand using the information-pushing method is smaller, operation
Speed faster, and then also accordingly promotes the corresponding speed of instruction.
In a kind of exemplary embodiment, step S14 specifically includes step S241 and step S242, referring to FIG. 2, Fig. 2
It is a kind of embodiment flow chart shown according to an exemplary embodiment using push probability pushed information.
In step S241, according to the corresponding push probability of information to be pushed, formation probability section.
For example, if the push probability of corresponding information to be pushed is 0.3, then the probability interval being correspondingly formed is [0,0.3].
In step S242, the random number of corresponding information to be pushed is generated.
Correspondingly, random number is generated according to the quantity of information to be pushed.In this example information to be pushed quantity be 1, then with
Machine number will be the number generated at random for being in [0,1].
In step S243, judge whether random number is in the corresponding probability interval.
In step S244, if random number is in the corresponding probability interval, by the information to be pushed of corresponding random number
It launches in accession page and is pushed, otherwise do not pushed.
For example, belong to probability interval [0,0.3] if the random number generated in the above process is 0.2, then it is eligible, it can
To execute the push of information to be pushed.If the random number generated in the above process is 0.5, it is not belonging to probability interval [0,0.3], then
It is eligible, the push of information to be pushed is not executed.
In the present embodiment, 1 information to be pushed is also greater than to corresponding number, the process of push is executed according to push probability
It is demonstrated.For example, information to be pushed quantity is 3.
In step S241, according to the corresponding push probability of above-mentioned information to be pushed, respectively 0.5,0.3,0.8.Wait push away
Push the sum of the probability of breath of delivering letters is less than the probability for necessarily pushing information to be pushed.By above-mentioned push probability according to from small to large
Sequence arranges, and is followed successively by 0.3,0.5,0.8.The push probability completed successively cumulative formation probability section will be arranged, be followed successively by [0,
0.3], (0.3,0.8], (0.8,1.6], each probability interval corresponds to information to be pushed.
In step S242, generate the random number of corresponding information to be pushed, in this example the value interval of random number according to
The quantity of information to be pushed is [0,3].According to random algorithm, the random number of generation is 1.5,0.2,1.8 respectively.
In step S243, judge whether random number is in the corresponding probability interval.
In step S244, if random number is in the corresponding probability interval, by the information to be pushed of corresponding random number
It launches in accession page and is pushed, otherwise do not pushed.
In above process, random number 1.5 be in probability interval (0.8,1.6], random number 0.2 is in [0,0.3], at random
Number 1.8 does not fall within corresponding probability interval.Judging result explanation, probability interval (0.8,1.6], [0,0.3] corresponding letter to be pushed
Breath will be pushed, i.e. the information to be pushed of push probability corresponding 0.3,0.8 can execute push.In addition, information to be pushed
Push can also be set, if push probability, which is greater than preset threshold, just executes above-mentioned steps.
In above process, the push of the information to be pushed is to present in accession page to user.In other possibility
Embodiment in, the information to be pushed can also remind the page, reading page, the page presentations such as welcome page.For example,
In accession page, there are two the positions to be pushed for push, then above-mentioned two pushed informations for meeting push request can
To randomly choose any position to be pushed, naturally it is also possible to specify position to be pushed according to other factors.If position to be pushed
Quantity is less than the quantity of qualified information to be pushed, then chooses the push biggish corresponding information to be pushed of probability.Some
In embodiment, a position to be pushed, which can also correspond to, accommodates multiple information to be pushed.
It further include step S351 and step S352 before step S14 in a kind of exemplary embodiment, referring to FIG. 3,
Fig. 3 is a kind of embodiment stream that information to be pushed amount shown according to an exemplary embodiment combines push probability pushed information
Cheng Tu.
In step S351, the maximum pushed information amount of accession page is obtained.
The accession page refers to the maximum pushed information amount that can carry in the page that information to be pushed is shown.It is described most
Big pushed information amount can correspond to the maximum pushed information amount of the individual access page, can also be for several access of push target
The maximum information push amount of the page.The accession page can be the accession page individually independently jumped, be also possible to seamless rank
The page flow connect.The page flow can continuously load access content, show information to be pushed and other information continuously
In accession page.The page flow can also set a certain period or a certain instruction, make information to be pushed on page flow and
Other information continuous updating.
In step S352, judge whether information to be pushed quantity is less than or equal to maximum pushed information amount.
If so, execution is described according to the corresponding push probability of information to be pushed, otherwise formation probability section does not push.
In the present embodiment, judge the numerical relation of information to be pushed quantity and maximum pushed information amount, and limited with this
Whether push is executed.If the overall quantity of information to be pushed alreadys exceed maximum pushed information amount, do not push.For changing
It, it is whether big with regard to no longer carrying out the push probability if the overall quantity of information to be pushed alreadys exceed maximum pushed information amount
In the judgement of preset threshold.
By taking the maximum pushed information amount of the individual access page as an example, i.e., maximum can permit push in the individual access page
Information to be pushed.In the present embodiment, maximum pushed information amount can push letter by counting the history of corresponding accession page
The factors such as breath amount, network quality are judged or are directly acquired artificial preset maximum pushed information amount.Assuming that individual access
The maximum pushed information amount of the page is 10, if amounting to 9 according to this push is counted, whether big can execute push probability
In the judgement of preset threshold；If 13 are amounted to according to this push is counted, then will no longer push.
In other possible embodiments, maximum pushed information can also be seen if fall out by the size of ordinal number
The limitation of amount.For example, the maximum pushed information amount of the individual access page be 5, the serial number 3 of information to be pushed, indicate if
Information to be pushed can be pushed, this article of information to be pushed is exactly the 3rd article of information to be pushed, reflect the information to be pushed number
Amount is less than maximum pushed information amount, can execute the judgement whether push probability is greater than preset threshold；If the sequence of information to be pushed
Number be 7, indicates if information to be pushed can be pushed, this article of information to be pushed is exactly the 7th article of information to be pushed, reflection institute
It states information to be pushed quantity and is greater than maximum pushed information amount, then do not execute the judgement whether push probability is greater than preset threshold.With
In the technical solution of serial number judgement, can also directly it judge after the serial number of information to be pushed is greater than maximum pushed information amount
It show that the accession page has reached maximum pushed information amount, is no longer pushed in this accession page.
In the scheme of page flow, maximum information push amount can be set by way of dividing virtual page number.Maximum letter
Push amount is ceased in addition to the individual access page can be directed to, and can also be configured for multiple accession pages.Other possible
In embodiment, maximum information push amount also can be set for specific push object, particular push object is avoided to receive
More pushed informations.In other possible embodiments, the maximum information push amount can be quantified as one of constraint item
Part participates in push computation model, restrains push computation model further.
The operand of the push computation model can be made to be further reduced by limiting maximum pushed information amount, accelerate fortune
Speed is calculated, further promotion response speed.Meanwhile it limiting maximum pushed information amount and can control shown in the accession page
Information avoids the ratio for occurring information to be pushed in access information excessively high.The limitation of maximum information push amount, can also limit to
The maximum of pushed information avoids pushed information from excessively causing unnecessary puzzlement.It is worth noting that,
Under the inspiration of the application, maximum information push amount specific value can adjust according to the actual situation.Possible
When situation or specific demand, the maximum information push amount can be set as being equal to the maximum information carried in accession page
Amount.
In a kind of exemplary embodiment, the push probability of the information to be pushed is calculated using the push computation model
Include:
minX(-Σi,j log(1+Aij Xij))；
Constraint condition is Σi Xij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate in i-th of accession page pushing away for j-th of information
Send probability, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
In the exemplary embodiment, the push computation model is under the constraint of push amount vector sum score matrix, can be with
Obtain push probability Xij.Push probability XijIt is expressed as the corresponding position of j-th of information in i-th of accession page, in the position
Set the push probability of corresponding information to be pushed.AijIndicate score matrix, i.e., j-th of the information in i-th of accession page is corresponding
Position, correspond to the score of information to be pushed in the position.djIndicate push amount vector, i.e., information to be pushed is in individual access page
Face is in push demand corresponding in the unit time.
For example, push probability X56It is expressed as the 6th corresponding position of information in the 5th accession page, in the position
The push probability of corresponding information to be pushed.A49Indicate score matrix, i.e., the 9th information in the 4th accession page is corresponding
Position corresponds to the score of information to be pushed in the position.In other embodiments, probability X is pushedijCorrespondence pushed information can
To share corresponding position to be pushed.
For example, dj=100 were expressed as in five minutes, and the corresponding push demand to pushing video is 100 times.Expression pair
It answers information to be pushed to need to be pushed 100 powers and is able to satisfy requirement.
Further, it is also possible to introduce Σi XijIt is limited less than or equal to number of positions to be pushed.For example, when in individual access page
In face, one is only provided with when pushing position for information to be pushed realization push, then introduces constraint Σi Xij≤1.Accordingly
Ground can also be set as the limitation of other quantity, not limited by the example above according to the limitation of number of positions to be pushed.Pass through
Σi XijWith the limitation wait push quantity between position, the quantity of feasible solution can be limited.In other embodiment, same
Fixed position to be pushed is provided in one page, all information to be pushed can share the position to be pushed, each wait push away
Send position that can also accommodate two or more pushed informations.
In the present embodiment, probability X is pushedij>=0, make algorithmic statement, excludes the solution for the push probability for lacking practical significance,
The operand of above formula is reduced, improving operational speed reduces operation time.
Push computation model in the present embodiment also applies, the operational form of matrix, may be implemented a plurality of wait push
Operation while information meets the push demand of a large amount of information to be pushed.Meanwhile the push computation model also applies convex function
Or the transformation of concave function, the solution of mating combination maximum value or minimum value can simplify operation, faster and more accurately pushed
Probability Xij.In above formula, also setting premise ΣiXij=dj, the sum of the push probability of premise expression in i accession page needs
Equal to push amount vector djRequirement, i.e. above formula includes to meet the limitation of push amount.
In the present embodiment, the limitation that may be incorporated into the maximum information push amount, forces algorithmic statement, keeps operation fast
Degree faster, solves more accurate.For example, being limited in the individual access page, a pushed information can be only pushed.
In another exemplary embodiment, the push for calculating the information to be pushed using the push computation model is general
Rate includes:
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is Σi Xij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djExpression pushes away
The amount of sending vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate the antithesis of push probability
Variable, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij
Expression formula calculate Xij。
In the present embodiment, the push computation model will push probability X by dualistic transformationijIt is converted to general with push
The relevant dual variable α of rate antithesisj.Solve dual variable αjSolve push probability Xij.Push probability XijIt is expressed as at i-th
The corresponding position of j-th of information in accession page, corresponds to the push probability of information to be pushed in the position.AijIndicate score square
Battle array, i.e., the corresponding position of j-th of information in i-th of accession page corresponds to the score of information to be pushed in the position.djTable
Show push amount vector, i.e., information to be pushed is in the individual access page in push demand corresponding in the unit time.Above-mentioned KKT item
Part refers to Karush-Kuhn-Tucher condition, and in optimum theory, KKT condition is the necessary item of Non-Linear Programming optimum solution
Part.KKT condition extends to the RegionAlgorithm for Equality Constrained Optimization in lagrange's method of multipliers (Lagrange multipliers) not
Equality constraint.
For example, push probability X13It is expressed as the 3rd corresponding position of information in the 1st accession page, in the position
The push probability of corresponding information to be pushed.A97Indicate score matrix, i.e., the 7th information in the 9th accession page is corresponding
Position corresponds to the score of information to be pushed in the position.Dual variable αjOperand relative to directly obtain push probability Xij
Operand it is smaller, arithmetic speed faster, and can obtain in real time push probability XijNumerical value, overcome accumulation certain amount or one
The problem of segment data of fixing time could solve makes above-mentioned push computation model can be used on line pushing above-mentioned information to be pushed.
Refer to a kind of online, push in real time on line therein.Correspondingly, refer under line need to store a certain amount of historical data with
Obtain push probability XijProcess.The historical data for the regular period that needs to rely on, the number within e.g. 10min are calculated under line
According to then calculating on line can be directly according to real-time score matrix AijRealize real-time operation.By solving dual variable αjIt can
Probability X is pushed to realizeijReal-time acquisition.
For in one embodiment, wherein inspiring constraint siRefer to maximum supply volume in accession page, i.e., maximum letter
Push amount is ceased, constraint s is inspirediIt is also denoted as si≤Σj Xij, the upper limit therein that can be understood as push quantity.Assuming that
siIt is set as 3, then it represents that more than the quantity of three information to be pushed, then no longer realize push.Such as when pushing to the 90th article extremely
When 100 information to be pushed, due to 90 to 100 exceeded inspiration constraint siThe upper limit, inspire constrain siLater to
Pushed information, it will make its corresponding push probability force zero setting, i.e., no longer realize push.
In the present embodiment, s is also introducediConstraint is inspired, the inspiration constraint representation is a kind of based on intuitive or experience
The constraint of construction obtains a feasible solution of each example of combinatorial optimization problem to be solved under acceptable corresponding conditions,
The departure degree of the feasible solution and optimal solution cannot be generally expected.In other possible embodiments, inspire constraint that can adopt
With the related constraint of imitative nature body, such as correspondence can be passed through using modes such as ant group algorithm, simulated annealing, neural networks
Algorithm obtain inspire constraint occurrence.It introduces and inspires constraint, reduce the operand of the pushed information model, pursuing more
Quick operation and under the premise of the shorter response time, can be applied.
In the push computation model, also setting premise ΣiXij=dj, premise expression pushing away in i accession page
The sum of probability is sent to need to be equal to push amount vector djRequirement, i.e. above formula includes to meet the limitation of push amount.In the present embodiment,
The limitation that may be incorporated into the maximum information push amount, prevents the quantity of information to be pushed excessive.For example, being limited in single visit
It asks in the page, confirmation only pushes 3 pushed informations.
It further include step S461, step S462 and step before step S11 in a kind of exemplary embodiment
S463, referring to FIG. 4, Fig. 4 is a kind of embodiment process shown according to an exemplary embodiment for obtaining information to be pushed
Figure.
In step S461, the pageview of pushed information on the page is obtained.
In the present embodiment, by way of crawl page access or the storage of other data, pageview is obtained.It is described clear
The amount of looking at may include the information such as page access amount, full page pageview, the pageview of page specifying information, click volume.It is described
Multimedia resource includes one or more multimedia messages presented with comprehensive form of video, audio, image, text.With view
For frequency, the pageview or click volume of video can be obtained by accession page itself, if there are multiple views in page access
Frequently, the pageview of corresponding video can also be obtained respectively.
In one embodiment, the operation of pageview is obtained, can be updated according to certain period, such as per minute,
It can also update according to certain trigger condition, such as user's touch-control more new button or gesture sliding and obtain pageview.
In step S462, the information to be pushed is extracted from the pushed information according to the pageview and is stored.
In the present embodiment, information to be pushed refers to the push unsatisfied information to be pushed of demand.To pageview point
Analysis, and the information to be pushed met to push request is extracted according to pageview.By taking video as an example, according to the sieve of video tour amount
Choosing obtains the push unsatisfied video of demand, if there are multiple videos for meeting condition in accession page, it can be together
It extracts.
It in one embodiment, can be extracted clear according to the either end of user, user terminal or server in real time
The amount of looking at obtains information to be pushed.For example, the information pushed in real time can be obtained by server, meet demand amount is detected whether.
In the present embodiment, according to the screening conditions of pageview, the information to be pushed that extraction obtains is stored, in order to subsequent applications.
In the present embodiment, can also when the corresponding push demand of information to be pushed or pageview acquisition meet, leave out storage to
Information to be pushed is transferred to the storage of other regions by pushed information, to reduce the pressure of corresponding storage, promotes storage efficiency.
In a kind of exemplary embodiment, the score matrix is used for characterization information clicking rate, the powder rate that rises, thumbs up rate, powder
The push effect of silk online hours or likability.In the present embodiment, the score matrix is by specifically pushing demand, and with
The corresponding push demand of a certain amount index characterization.For example, certain information to be pushed is needed with the push that the rate that thumbs up reaches 80%
It asks, converts corresponding score for the rate demand of thumbing up, and by fraction set ingredient matrix number, the corresponding push of score matrix mapping
The score of information.In addition, score matrix can characterize the demand of push number of users, it is e.g. within the unit time or certain
In the accession page of quantity, the demand of information to be pushed is pushed to the user of preset quantity.
In the present embodiment, the authenticity or active degree of bean vermicelli online hours reflection bean vermicelli, can avoid to a certain degree
The cheating mode of virtual bean vermicelli or imaginary data promotes the authenticity that demand is satisfied.The likability reflects user couple
The degrees of tolerance of information to be pushed, for example, user for information to be pushed close, complain, loseing interest in etc. it is at least one feed back,
Can determine user to information to be pushed and lose interest in or detest, the strategy of adjustable push at this time.It is other can
In the embodiment of energy, the score in the score matrix can be the composite score of a variety of push effects, for a variety of push
Effect can be counted by way of centainly weighting passes through the combined technical solution of this weighting for reciprocal fraction, it is possible to reduce
Push the foundation of computation model, it is possible to reduce operand.It is of course also possible to be built for a variety of push effect combination push amount vectors
Found multiple push computation models.
In a kind of exemplary embodiment, the information to be pushed includes at least one of video, audio, image, text
Information.In the present embodiment, the information to be pushed can be video, i.e., the object of the described information-pushing method push is view
Frequently, its object is to realize the push for making video obtain effect of optimization under the premise of meeting and pushing demand.
For example, can use the information-pushing method when having popularization demand to certain video or certain batch of video, guarantee
In the case that push amount obtains satisfaction, the push effect of optimization, e.g. clicking rate, user's rate of people logging in, comment rate, broadcasting are obtained
Amount etc..In other possible embodiments, information to be pushed is it is also possible that text, the text further include being able to carry out to jump
Link, described be connected in other modes in the cards can be combined with the display forms such as picture, video, enriches and shows shape
Formula.
In a kind of exemplary embodiment, other limitations can also be added to according to the push probability pushed information,
Such as according to information such as geographical location information, age information, interactive information, subject information, history pushed informations, push is carried out
Limitation.By taking geographical location information as an example, for Pekinese user, information to be pushed relevant to Beijing can be pushed；For wide
It is relevant to PUSH message can to push Guangdong,Hongkong and Macao area by the user in state.By taking interactive information as an example, user good friend quilt can be directed to
The information of push can push identical or approximate information, and preferably push effect is realized based on social activity.In the present embodiment
In, by adding above-mentioned restricted information, the push of information to be pushed can be made more accurate, promote the accuracy of push effect.
Fig. 5 is a kind of block diagram of information push-delivery apparatus shown according to an exemplary embodiment.Referring to Fig. 5, the device packet
Include analytical unit 51, modeling unit 52, arithmetic element 53 and push unit 54:
Analytical unit 51 is configured as establishing score matrix and push amount vector according to information to be pushed, the score square
Battle array is the matrix of component number building by information push effect quantization；
Modeling unit 52 is configured as establishing push computation model according to the score matrix and push amount vector；Wherein,
The push computation model be under the constraint condition for meeting the score matrix and push amount vector, what pushed information was pushed
The computation model of parameter probability valuing；
Arithmetic element 53 is configured as calculating the push probability of the information to be pushed using push computation model；
Push unit 54 is configured as pushing the information to be pushed according to the push probability.
Fig. 6 is the block diagram of another information push-delivery apparatus shown according to an exemplary embodiment, is to show that section is raw
At unit 641, random number generation unit 642, random number judging unit 643 and execute push unit 644.Please refer to Fig. 6, In
In the present embodiment push unit 54 include section generation unit 641, random number generation unit 642, random number judging unit 643 with
And execute push unit 644:
Section generation unit 641 is configured as according to the corresponding push probability of information to be pushed, formation probability section.
Random number generation unit 642 is configurable to generate the random number of corresponding information to be pushed.
Random number judging unit 643 is configured as judging whether random number is in the corresponding probability interval.
Push unit 644 is executed, if being configured as random number is in the corresponding probability interval, by corresponding random number
Information to be pushed dispensing is pushed into accession page, is not otherwise pushed.
Fig. 7 is the block diagram of another information push-delivery apparatus shown according to an exemplary embodiment, is to show that first obtains
Take unit 751, information content judging unit 752, section generation unit 741, random number generation unit 742, random number judging unit
743 and execute push unit 744.Referring to Fig. 7, the information push-delivery apparatus further includes first acquisition unit in the present embodiment
751, information content judging unit 752, section generation unit 741, random number generation unit 742, random number judging unit 743 and
Execute push unit 744:
First acquisition unit 751 is configured as obtaining the maximum pushed information amount of the individual access page；
Information content judging unit 752 is configured as judging whether information to be pushed quantity is less than or equal to maximum push letter
Breath amount；
Section generation unit is configured as if so, according to the corresponding push probability of information to be pushed, formation probability section,
Otherwise it does not push；
Random number generation unit 742 is configurable to generate the random number of corresponding information to be pushed；
Random number judging unit 743 is configured as judging whether random number is in the corresponding probability interval；
Push unit 744 is executed, if being configured as random number is in the corresponding probability interval, by corresponding random number
Information to be pushed dispensing is pushed into accession page, is not otherwise pushed.
In a kind of exemplary embodiment, the push probability of the information to be pushed is calculated using the push computation model
Include:
minX(-Σi,j log(1+Aij Xij))；
Constraint condition is Σi Xij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate in i-th of accession page pushing away for j-th of information
Send probability, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
In another exemplary embodiment, the push for calculating the information to be pushed using the push computation model is general
Rate includes:
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is Σi Xij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djExpression pushes away
The amount of sending vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate the antithesis of push probability
Variable, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij
Expression formula calculate Xij。
In a kind of exemplary embodiment, the score matrix is used for characterization information clicking rate, the powder rate that rises, thumbs up rate, powder
The push effect of silk online hours or likability.
In a kind of exemplary embodiment, the information to be pushed includes at least one of video, audio, image, text
Information.
Fig. 8 is the block diagram of another information push-delivery apparatus shown according to an exemplary embodiment, is to show that second obtains
Take unit 861 and extraction unit 862.Referring to Fig. 8, the information push-delivery apparatus further includes second acquisition unit in the present embodiment
861, extraction unit 862 and storage unit 863:
Second acquisition unit 861 is configured as obtaining the pageview of pushed information on the page；
Extraction unit 862 is configured as extracting the information to be pushed from the pushed information according to the pageview
And it stores.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method
Embodiment in be described in detail, no detailed explanation will be given here.
In one embodiment, it is also proposed that a kind of Information Push Server, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to the step of the described information-pushing method.
It in the present embodiment, can also include the equipment, e.g. hand such as mobile communication equipment, computer in addition to server
The electronic equipments such as machine, plate, notebook.
About the Information Push Server in above-described embodiment, the concrete mode that wherein modules execute operation exists
It is described in detail in embodiment in relation to this method, no detailed explanation will be given here.
In one embodiment, it is also proposed that a kind of non-transitorycomputer readable storage medium, when the storage medium
In instruction when being executed by the processor of server so that non-transitorycomputer readable storage medium is able to carry out described push away
Send information approach.
About the non-transitorycomputer readable storage medium in above-described embodiment, wherein each concrete mode is related
It is described in detail in the embodiment of this method, no detailed explanation will be given here.
Fig. 9 is a kind of block diagram of device 900 for information push shown according to an exemplary embodiment.For example, dress
Setting 900 can be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical treatment
Equipment, body-building equipment, personal digital assistant etc..
Referring to Fig. 9, device 900 may include following one or more components: processing component 902, memory 904, electric power
Component 906, multimedia component 908, audio component 910, the interface 912 of input/output (I/O), sensor module 914, and
Communication component 916.
The integrated operation of the usual control device 900 of processing component 902, such as with display, telephone call, data communication, phase
Machine operation and record operate associated operation.Processing component 902 may include that one or more processors 920 refer to execute
It enables, to perform all or part of the steps of the methods described above.In addition, processing component 902 may include one or more modules, just
Interaction between processing component 902 and other assemblies.For example, processing component 902 may include multi-media module, it is more to facilitate
Interaction between media component 908 and processing component 902.
Memory 904 is configured as storing various types of data to support the operation in equipment 900.These data are shown
Example includes the instruction of any application or method for operating on device 900, contact data, and telephone book data disappears
Breath, picture, video etc..Memory 904 can be by any kind of volatibility or non-volatile memory device or their group
It closes and realizes, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable to compile
Journey read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash
Device, disk or CD.
Power supply module 906 provides electric power for the various assemblies of device 900.Power supply module 906 may include power management system
System, one or more power supplys and other with for device 900 generate, manage, and distribute the associated component of electric power.
Multimedia component 908 includes the screen of one output interface of offer between described device 900 and user.One
In a little embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touch sensings
Device is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding action
Boundary, but also detect duration and pressure associated with the touch or slide operation.In some embodiments, more matchmakers
Body component 908 includes a front camera and/or rear camera.When equipment 900 is in operation mode, such as screening-mode or
When video mode, front camera and/or rear camera can receive external multi-medium data.Each front camera and
Rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 910 is configured as output and/or input audio signal.For example, audio component 910 includes a Mike
Wind (MIC), when device 900 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone is matched
It is set to reception external audio signal.The received audio signal can be further stored in memory 904 or via communication set
Part 916 is sent.In some embodiments, audio component 910 further includes a loudspeaker, is used for output audio signal.
I/O interface 912 provides interface between processing component 902 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lock
Determine button.
Sensor module 914 includes one or more sensors, and the state for providing various aspects for device 900 is commented
Estimate.For example, sensor module 914 can detecte the state that opens/closes of equipment 900, and the relative positioning of component, for example, it is described
Component is the display and keypad of device 900, and sensor module 914 can be with 900 1 components of detection device 900 or device
Position change, the existence or non-existence that user contacts with device 900,900 orientation of device or acceleration/deceleration and device 900
Temperature change.Sensor module 914 may include proximity sensor, be configured to detect without any physical contact
Presence of nearby objects.Sensor module 914 can also include optical sensor, such as CMOS or ccd image sensor, at
As being used in application.In some embodiments, which can also include acceleration transducer, gyro sensors
Device, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 916 is configured to facilitate the communication of wired or wireless way between device 900 and other equipment.Device
900 can access the wireless network based on communication standard, such as WiFi, carrier network (such as 2G, 3G, 4G or 5G) or them
Combination.In one exemplary embodiment, communication component 916 is received via broadcast channel from the wide of external broadcasting management system
Broadcast signal or broadcast related information.In one exemplary embodiment, the communication component 916 further includes near-field communication (NFC)
Module, to promote short range communication.For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) can be based in NFC module
Technology, ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 900 can be believed by one or more application specific integrated circuit (ASIC), number
Number processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided
It such as include the memory 904 of instruction, above-metioned instruction can be executed by the processor 920 of device 900 to complete the information push side
Method.For example, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic
Band, floppy disk and optical data storage devices etc..
Figure 10 is a kind of block diagram of device 1000 for information-pushing method shown according to an exemplary embodiment.Example
Such as, device 1000 may be provided as a server.Referring to Fig.1 0, device 1000 includes processing component 1022, is further wrapped
One or more processors, and the memory resource as representated by memory 1032 are included, it can be by processing component for storing
The instruction of 1022 execution, such as application program.The application program stored in memory 1032 may include one or one with
On each correspond to one group of instruction module.In addition, processing component 1022 is configured as executing instruction, to execute the letter
Cease method for pushing.
Device 1000 can also include that a power supply module 1026 be configured as the power management of executive device 1000, and one
Wired or wireless network interface 1050 is configured as device 1000 being connected to network and input and output (I/O) interface
1058.Device 1000 can be operated based on the operating system for being stored in memory 1032, such as Windows ServerTM, Mac
OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
In one embodiment, a kind of computer program product, including computer program code, the calculating are further related to
Machine program includes program instruction, when described program instruction is computer-executed, the computer is made to execute the information push
Method.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the application
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and including the undocumented common knowledge in the art of the application
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim is pointed out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (18)
1. a kind of information-pushing method characterized by comprising
Establish score matrix and push amount vector according to information to be pushed, the score matrix be by information push effect quantization and
The matrix of component number building；
Push computation model is established according to the score matrix and push amount vector；Wherein, the push computation model is full
The computation model for the parameter probability valuing that pushed information is pushed under the constraint condition of the foot score matrix and push amount vector；
The push probability of the information to be pushed is calculated using push computation model；
The information to be pushed is pushed according to the push probability.
2. information-pushing method according to claim 1, which is characterized in that described wait push away according to push probability push
It delivers letters breath, comprising:
According to the corresponding push probability of information to be pushed, formation probability section；
Generate the random number of corresponding information to be pushed；
Judge whether random number is in the corresponding probability interval；
If random number is in the corresponding probability interval, by the information to be pushed of corresponding random number launch into accession page into
Row push, does not otherwise push.
3. information-pushing method according to claim 2, which is characterized in that judge described corresponding according to information to be pushed
Probability is pushed, before formation probability section, further includes:
Obtain the maximum pushed information amount of accession page；
Judge whether information to be pushed quantity is less than or equal to maximum pushed information amount；
If so, according to information to be pushed corresponding push probability otherwise formation probability section does not push.
4. information-pushing method according to claim 1, which is characterized in that using described in push computation model calculating
The push probability of information to be pushed includes:
minX(-Σi,jlog(1+AijXij))；
Constraint condition is ΣiXij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate that the push of j-th of information in i-th of accession page is general
Rate, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
5. information-pushing method according to claim 4, which is characterized in that using described in push computation model calculating
The push probability of information to be pushed includes:
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is ΣiXij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djIndicate push amount
Vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate push probability to mutation
Amount, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij's
Expression formula calculates Xij。
6. information-pushing method according to claim 2, which is characterized in that described to establish score square according to information to be pushed
Before battle array and push amount vector, the information-pushing method further include:
Obtain the pageview of pushed information on the page；
The information to be pushed is extracted from the pushed information according to the pageview and is stored.
7. information-pushing method according to claim 1, which is characterized in that the score matrix is clicked for characterization information
Rate, the powder rate that rises, the push effect for thumbing up rate, bean vermicelli online hours or likability.
8. information-pushing method according to claim 1, which is characterized in that the information to be pushed include video, audio,
At least one information of image, text.
9. a kind of information push-delivery apparatus, which is characterized in that the information push-delivery apparatus includes:
Analytical unit is configured as establishing score matrix and push amount vector according to information to be pushed, the score matrix be by
Information pushes the matrix of effect quantization and component number building；
Modeling unit is configured as establishing push computation model according to the score matrix and push amount vector；Wherein, described to push away
Sending computation model is that the probability that pushed information is pushed under the constraint condition for meeting the score matrix and push amount vector takes
The computation model of value；
Arithmetic element is configured as calculating the push probability of the information to be pushed using push computation model；
Push unit is configured as pushing the information to be pushed according to the push probability.
10. information push-delivery apparatus according to claim 9, which is characterized in that the push unit includes:
Section generation unit is configured as according to the corresponding push probability of information to be pushed, formation probability section；
Random number generation unit is configurable to generate the random number of corresponding information to be pushed；
Random number judging unit is configured as judging whether random number is in the corresponding probability interval；
Push unit is executed, if being configured as random number is in the corresponding probability interval, by corresponding random number wait push
Information dispensing is pushed into accession page, is not otherwise pushed.
11. information push-delivery apparatus according to claim 9, which is characterized in that the information push-delivery apparatus further include:
First acquisition unit is configured as obtaining the maximum pushed information amount of accession page；
Information content judging unit is configured as judging whether information to be pushed quantity is less than or equal to maximum pushed information amount；
Section generation unit is configured as if so, according to the corresponding push probability of information to be pushed, formation probability section, otherwise
It does not push.
12. information push-delivery apparatus according to claim 9, which is characterized in that calculate institute using the push computation model
The push probability for stating information to be pushed includes:
minX(-Σi,jlog(1+AijXij))；
Constraint condition is ΣiXij=dj；
Xij≥0；
Wherein, AijIndicate that score matrix, X indicate push matrix, XijIndicate that the push of j-th of information in i-th of accession page is general
Rate, djIndicate that push amount vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page.
13. information push-delivery apparatus according to claim 12, which is characterized in that calculate institute using the push computation model
The push probability for stating information to be pushed includes:
Using KKT condition, in conjunction with heuristic constraint, by XijIt is expressed as following formula:
Xij=min (si,max(0,1/αj–1/Aij))；
Constraint condition is ΣiXij=dj；
Wherein, AijIndicate score matrix, XijIndicate the push probability of j-th of information in i-th of accession page, djIndicate push amount
Vector, i indicate that accession page ordinal number, j indicate the information ordinal number in the individual access page, αjIndicate push probability to mutation
Amount, siIt indicates to inspire constraint, KKT condition indicates the necessary condition of Non-Linear Programming optimum solution；Solve αj, pass through above-mentioned Xij's
Expression formula calculates Xij。
14. information push-delivery apparatus according to claim 10, which is characterized in that the information push-delivery apparatus further include:
Second acquisition unit is configured as obtaining the pageview of pushed information on the page；
Extraction unit is configured as extracting the information to be pushed from the pushed information according to the pageview and store.
15. information push-delivery apparatus according to claim 9, which is characterized in that the score matrix is used for characterization information point
Hit rate, the powder rate that rises, the push effect for thumbing up rate, bean vermicelli online hours or likability.
16. information push-delivery apparatus according to claim 9, which is characterized in that the information to be pushed includes video, sound
Frequently, at least one information of image, text.
17. a kind of Information Push Server characterized by comprising
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to: the information-pushing method as described in any one of claims 1 to 8 claim
The step of.
18. a kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by the processor of server
When execution, enable the server to execute the pushed information method as described in any one of claims 1 to 8 claim.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811117810.XA CN109379410B (en) | 2018-09-21 | 2018-09-21 | Information-pushing method, device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811117810.XA CN109379410B (en) | 2018-09-21 | 2018-09-21 | Information-pushing method, device, server and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109379410A CN109379410A (en) | 2019-02-22 |
| CN109379410B true CN109379410B (en) | 2019-11-19 |
Family
ID=65402463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811117810.XA Active CN109379410B (en) | 2018-09-21 | 2018-09-21 | Information-pushing method, device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109379410B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619079B (en) * | 2019-06-20 | 2022-08-23 | 北京无限光场科技有限公司 | Method and device for pushing information |
| CN112150182B (en) * | 2019-06-28 | 2023-08-29 | 腾讯科技（深圳）有限公司 | Multimedia file pushing method and device, storage medium and electronic device |
| CN110740166B (en) * | 2019-09-19 | 2022-06-17 | 平安科技（深圳）有限公司 | Distance-based information sending method and device, computer equipment and storage medium |
| CN114610022B (en) * | 2022-02-09 | 2024-07-02 | 北京三快在线科技有限公司 | Connection positioning method and distribution system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105183904B (en) * | 2015-09-30 | 2020-01-10 | 北京金山安全软件有限公司 | Information pushing method and device and electronic equipment |
| CN105260477A (en) * | 2015-11-06 | 2016-01-20 | 北京金山安全软件有限公司 | Information pushing method and device |
| US20180096420A1 (en) * | 2016-10-05 | 2018-04-05 | Aiooki Limited | Enhanced Bidding System |
| CN108399551A (en) * | 2017-02-08 | 2018-08-14 | 阿里巴巴集团控股有限公司 | A kind of method and system of determining user tag and pushed information |
| CN108304440B (en) * | 2017-11-01 | 2021-08-31 | 腾讯科技（深圳）有限公司 | Game pushing method and device, computer equipment and storage medium |
| CN109902222B (en) * | 2018-11-30 | 2022-05-13 | 华为技术有限公司 | Recommendation method and device |
-
2018
- 2018-09-21 CN CN201811117810.XA patent/CN109379410B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109379410A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109379410B (en) | Information-pushing method, device, server and storage medium | |
| CN110585726B (en) | User recall method, device, server and computer readable storage medium | |
| CN105431813B (en) | It is acted based on biometric identity home subscriber | |
| KR20190056538A (en) | Server and operating method thereof | |
| CN103944804B (en) | Contact recommending method and device | |
| CN110168586A (en) | Situation generates and the selection of the media content to customization | |
| US10726087B2 (en) | Machine learning system and method to identify and connect like-minded users | |
| CN105100197B (en) | The method and device of application is installed | |
| KR20230022983A (en) | Messaging system with external-resource dock and drawer | |
| US20210409354A1 (en) | Method and system for providing relevance information between users | |
| US11790565B2 (en) | Compressing image-to-image models with average smoothing | |
| CN106310641A (en) | Motion scheme recommendation method and device | |
| KR20240090286A (en) | Speech recognition with intent estimation | |
| CN114154068A (en) | Media content recommendation method and device, electronic equipment and storage medium | |
| CN110490389A (en) | Clicking rate prediction technique, device, equipment and medium | |
| KR20230160930A (en) | User presence display data management | |
| CN107368998A (en) | Agenda managing method and Related product | |
| CN111191143B (en) | Application recommendation method and device | |
| CN112256976B (en) | Matching method and related device | |
| KR20230124067A (en) | Compress image-to-image models | |
| CN113822435B (en) | Prediction method of user conversion rate and related equipment | |
| CN108829234A (en) | It is capable of the movement technique and system of online interaction | |
| CN113761784A (en) | Data processing method, training method and device of data processing model | |
| CN111898019A (en) | Information pushing method and device | |
| US11894989B2 (en) | Augmented reality experience event metrics system |
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 |