CN109214847A - Page clicking rate data processing method, apparatus and system - Google Patents
Page clicking rate data processing method, apparatus and system Download PDFInfo
- Publication number
- CN109214847A CN109214847A CN201710542895.5A CN201710542895A CN109214847A CN 109214847 A CN109214847 A CN 109214847A CN 201710542895 A CN201710542895 A CN 201710542895A CN 109214847 A CN109214847 A CN 109214847A
- Authority
- CN
- China
- Prior art keywords
- page
- clicking rate
- information
- probability
- history
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
Abstract
The object of the present invention is to provide a kind of page clicking rate data processing methods, apparatus and system, and wherein method is the following steps are included: the step of S1 is used to obtain the page clicking rate information of the different pages respectively；S2 is used to be analyzed and processed the page clicking rate information of the different pages of acquisition and generates the step of recommending page results information；S3 is for exporting the step of recommending page results information.The page clicking rate information of the available different pages of the present invention and maximum clicking rate result information, advertisement announcement putting person does not need this numerical value of click for manually counting each advertisement display page and artificial judgment provides the advertisement display page that rate potentiality are hit in highest point, to which selection launches advertisement in the page, thus time-consuming for the workload and flower that advertisement putting person can be reduced, the efficiency and accuracy for improving decision meet the demand for the decision that advertisement putting person is efficient, the convenient, advertisement that accurately makes a choice shows the page.
Description
Technical field
The present invention relates to artificial intelligence field more particularly to a kind of page clicking rate data processing methods, apparatus and system.
Background technique
Big data (big data, mega data) or flood tide data, refer to needing new tupe that could have
Stronger decision edge, the magnanimity of insight and process optimization ability, high growth rate and diversified information assets.In this big number
According to epoch, people mostly use all data to be analyzed and processed.Artificial intelligence is " artificial " and " intelligence ", " artificial " i.e. manpower
Reached manufacture, the degree of intelligence of people itself is either with or without high to the stage that can create artificial intelligence." intelligence " is related to all
Such as consciousness (CONSCIOUSNESS), self (SELF), thinking (MIND) (including unconscious thinking (UNCONSCIOUS_
)) etc. MIND the problems such as, in brief, the ability of decision is intelligently referred to.The research of artificial intelligence often relates to the intelligence to people
The research of itself is simultaneously found broad application in computer field.It is different in the practice that advertisement shows page selection
Show that its clicking rate of the page is different, the higher advertisement of clicking rate shows that the page more receives an acclaim, and advertisement putting person wishes numerous
Advertisement show that advertisement with highest clicking rate potentiality is selected in the page shows the page.In the prior art, it needs manually to unite
It counts each advertisement and shows this numerical value of the click of the page, and artificial judgment provides the advertisement display page that rate is hit in highest point, work
Amount is big, time-consuming, and accuracy and efficiency are lower, and it is aobvious to be unable to satisfy efficient, convenient, the accurate advertisement that makes a choice of advertisement putting person
Show the demand of the decision of the page.
Summary of the invention
The object of the present invention is to provide a kind of page clicking rate data processing methods, apparatus and system.
Page clicking rate data processing method provided by the present invention, comprising the following steps:
S1 is used to obtain the step of page clicking rate information of the different pages respectively；
S2 is used to be analyzed and processed the page clicking rate information of the different pages of acquisition and generates recommendation page results
The step of information；
S3 is for exporting the step of recommending page results information.
Page clicking rate data processing equipment provided by the present invention, comprising:
For obtaining the module of the page clicking rate information of the different pages respectively；
Page clicking rate information for the different pages to acquisition, which is analyzed and processed and generates, recommends page results letter
The module of breath；
For exporting the module for recommending page results information.
Page clicking rate data system provided by the present invention, including client, the client include: micro process control
Device, display screen and touch screen；The display screen and touch screen with the microprocessor controller circuit connection.It is provided by the present invention
Page clicking rate data processing method, apparatus and system, through this embodiment provided by page clicking rate data processing side
The page clicking rate information of the available different pages of method and recommendation page results information, advertisement putting person do not need manually to count
Each advertisement shows that this numerical value of click of the page and artificial judgment provide the advertisement display page that rate potentiality are hit in highest point, energy
The page appropriate is recommended by this method, device and processor, so that selection launches advertisement in the page, thus can reduce advertisement throwing
Time-consuming for the workload and flower for the person of putting, and improves the efficiency and accuracy of decision, it is efficient, convenient, accurate to meet advertisement putting person
The advertisement that makes a choice shows the demand of the decision of the page.
Detailed description of the invention
Fig. 1 is the flow diagram of page clicking rate data processing method of the present invention；
Fig. 2 is the specific steps figure of page clicking rate data processing method described in the embodiment of the present invention one；
Fig. 3 is page clicking rate data processing system client terminal structure schematic diagram described in the embodiment of the present invention three.；Fig. 4
For page clicking rate Data processing system network structural schematic diagram described in the embodiment of the present invention three.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Embodiment one
As shown in Figure 1, page clicking rate data processing method provided by the present embodiment, comprising the following steps:
S1 is used to obtain the step of page clicking rate information of the different pages respectively；
S2 is used to be analyzed and processed the page clicking rate information of the different pages of acquisition and generates recommendation page results
The step of information；
S3 is for exporting the step of recommending page results information.
The page clicking rate information, the number of clicks information of display number information and the page including the page.
It is described that recommend the page be the page clicking rate data processing method that is provided through the invention according to user demand more
The page for being suitble to the user demand that decision goes out in a page, the recommendation page described in the present embodiment are latent with maximum clicking rate
The page of power.
The information user can easily operate so as to show the recommendation page according to the recommendation page results information
Data or any operation is carried out without user according to the information, display equipment can show the letter for recommending the page automatically
Breath.
It will be understood to those skilled in the art that page clicking rate data processing method can provided by through this embodiment
To obtain the page clicking rate information of the different pages and recommend page results information, advertisement putting person does not need manually to count each
Advertisement shows that this numerical value of click of the page and artificial judgment provide the advertisement display page that rate is hit in highest point, can have maximum
Thus the page of clicking rate potentiality can reduce the workload and flower consumption of advertisement putting person so that selection launches advertisement in the page
Duration improves the efficiency and accuracy of decision, meets efficient, convenient, the accurate advertisement that makes a choice of advertisement putting person and shows page
The demand of the decision in face.
Further, the S1 be used for obtain respectively the different pages the page clicking rate information the step of include:
S11 sets the apriori probability distribution value information of each page clicking rate；
S12 obtains the history page clicking rate information of each page.
It will be understood to those skilled in the art that the apriori probability Distribution Value information includes Probability Distribution model in advance
(distribution model) and in advance Probability Distribution parameter (parameter), user can freely set the machine in advance
Rate is distributed value information.Such as in actual motion, when the model of Probability Distribution in advance for numbering the page clicking rate for being i divides in Gauss
Cloth (Gaussian distribution), Probability Distribution parameter is average value mu in advanceiAnd standard deviation sigmaiWhen (> 0), probability density
Function is
The history page clicking rate information, the click of display number history information and the page including the page
Number history information.
Further, S2 is used to be analyzed and processed the page clicking rate information of the different pages of acquisition and generates-recommend
The step of page results information includes:
S21 is distributed value information according to the history page clicking rate information and the apriori probability that obtain each page, calculates
The posterior probability distributed intelligence of page-out clicking rate；
S22 is clicked according to history and is recorded, and calculates the display number accounting of each page；
S23 calculates the winning rate of each page according to the posterior probability distributed intelligence；It will be understood by those skilled in the art that
The winning rate of the page refers to that the page becomes the percentage of winning for recommending the page.
S24 calculates the display number accounting and the winning rate posterior probability point of each page obtained according to above-mentioned steps
Cloth information, which calculates, recommends page results information.
It will be understood by those skilled in the art that through the above steps can be to the apriori probability of each page clicking rate got
The history page clicking rate information of distribution value information and each page is handled, and recommends page results letter to obtain
Breath, advertisement putting person can go out be suitble to the page of the user demand by the recommendation page results information decision, specific to this implementation
Example, i.e., decision provides the page that maximum point hits rate potentiality.To which selection launches advertisement in the page, advertisement throwing thus can be reduced
Time-consuming for the workload and flower for the person of putting, and improves the efficiency and accuracy of decision, it is efficient, convenient, accurate to meet advertisement putting person
The advertisement that makes a choice shows the demand of the decision of the page.
Further, value information is distributed according to the history page clicking rate information and the apriori probability that obtain each page,
The step of calculating the posterior probability distributed intelligence of page-out clicking rate uses the side of bayesian reasoning (Bayesian inference)
Method is calculated, to obtain the posterior probability distributed intelligence of page-out clicking rate.
Further, the method that the method with bayesian reasoning (Bayesian inference) is calculated are as follows:
' wherein, θ is that Probability Distribution parameter (distribution parameter), p (θ) they are apriori probability distribution in advance
(prior distribution), p (x | θ ') it is similar function (likelihood function), p (θ | x) posterior probability point
Cloth (posterior distribution).
Posterior probability distributed intelligence is it will be understood to those skilled in the art that with bayesian reasoning (Bayesian
Inference the posterior probability distributed intelligence for) calculating each page clicking rate reduces advertisement putting person to this numerical value of click and highest
The advertisement of clicking rate shows the artificial statistics and artificial judgment of the page, reduces cost.
Further, in the step of S22 is clicked according to history and recorded, and calculates the display number accounting of each page, work as volume
The cumulative number accounting value that number page for being i is shown is bi, the sample size (sampling for the history page display information that number is j
Number) it is njWhen, it is calculated in the following way；
Further, when the probability density function of the posterior probability distributed intelligence is fj(x), the posterior probability distribution letter
The probability cumulative density function of breath is Fj(x) when, relationship between the two is
Further, the probability density function of the posterior probability distributed intelligence for the page for being i when number
(probability density function, p.d.f.) is fi(x), the posterior probability point for the page for being j when number
The probability cumulative density function (cumulated density function, c.d.f.) of cloth information is Fj(x) and number is i
The page the winning rate piBetween relationship be
Further, the recommendation page number is k=argmax (pi-bi).In other words, the winning rate p of recommendation page kkWith
The accumulative display number accounting b of the recommendation pagekThe difference p of the twok-bkGreater than the winning rate p of other pages iiWith the page
Accumulative display number accounting biThe difference p of the twoi-bi。
It will be understood by those skilled in the art that may make each page winning rate p by the above methodiWith cumulative number accounting value
biMatch to the greatest extent.
Further, the S3 includes: for the step of exporting maximum clicking rate result information
Page results are recommended in S31 display；
S32 user selects the corresponding display page according to the recommendation page results of display and generates new history page click
Rate information；
S33 return step S12.
It will be understood to those skilled in the art that being fed back by new history page clicking rate information, the present embodiment is mentioned
The page clicking rate data processing equipment of confession can form self-study mechanism, accumulate new history page clicking rate information, this field
It is latent that technical staff is appreciated that the huger recommendation page results to acquisition of the data of history page clicking rate information more have
Power.
Embodiment two
Page clicking rate data processing equipment provided by the present embodiment, comprising:
For obtaining the module of the page clicking rate information of the different pages respectively；
Page clicking rate information for the different pages to acquisition, which is analyzed and processed and generates, recommends page results letter
The module of breath；
For exporting the module for recommending page results information.
The page clicking rate information, the number of clicks information of display number information and the page including the page.
It is described that recommend the page be the page clicking rate data processing method that is provided through the invention according to user demand more
The page for being suitble to the user demand that decision goes out in a page, the recommendation page described in the present embodiment are latent with maximum clicking rate
The page of power.
The information user can easily operate so as to show the recommendation page according to the recommendation page results information
Data or any operation is carried out without user according to the information, display equipment can show the letter for recommending the page automatically
Breath.
It will be understood to those skilled in the art that page clicking rate data processing equipment can provided by through this embodiment
It accuses putting person to obtain page clicking rate information and maximum clicking rate result information, the advertisement of the different pages and does not need manually to count
Each advertisement shows that this numerical value of click of the page and artificial judgment provide the advertisement display page that rate is hit in highest point, can learn
The clicking rate of which page is maximum, so that selection launches advertisement in the page, thus can reduce advertisement putting person workload and
Time-consuming for flower, improves the efficiency and accuracy of decision, and it is aobvious to meet efficient, convenient, the accurate advertisement that makes a choice of advertisement putting person
Show the demand of the decision of the page.
Further, the module of the page clicking rate information for obtaining the different pages respectively includes:
Apriori probability for setting each page clicking rate is distributed the submodule of value information；
For obtaining the submodule of the history page clicking rate information of each page.
It will be understood to those skilled in the art that the apriori probability Distribution Value information includes Probability Distribution model in advance
(distribution model) and in advance Probability Distribution parameter (parameter), user can freely set the machine in advance
Rate is distributed value information.Such as in actual motion, when the model of Probability Distribution in advance for numbering the page clicking rate for being i divides in Gauss
Cloth (Gaussian distribution), Probability Distribution parameter is average value mu in advanceiAnd standard deviation sigmaiWhen (> 0), probability density
Function is
The history page clicking rate information, the click of display number history information and the page including the page
Number history information.
Further, it is analyzed and processed for the page clicking rate information of the different pages to acquisition and generates the recommendation page
The module of result information includes:
For being distributed value information according to the history page clicking rate information and the apriori probability that obtain each page, calculate
The submodule of the posterior probability distributed intelligence of page-out clicking rate；It is recorded for being clicked according to history, calculates the display of each page
The submodule of number accounting；
For calculating the winning rate of each page according to the posterior probability distributed intelligence；It will be understood by those skilled in the art that
The winning rate of the page refers to that the page becomes the submodule for recommending the percentage of winning of the page.
The display number accounting of each page for being obtained according to above-mentioned steps calculating and the winning rate posterior probability
Distributed intelligence calculates the submodule for recommending page results information.
It will be understood by those skilled in the art that can be to the machine in advance of each page clicking rate got by above-mentioned submodule
The history page clicking rate information of rate distribution value information and each page is handled, and recommends page results letter to obtain
Breath, advertisement putting person can go out be suitble to the page of the user demand by the recommendation page results information decision, specific to this implementation
Example, i.e., decision provides the page that maximum point hits rate potentiality.To which selection launches advertisement in the page, advertisement throwing thus can be reduced
Time-consuming for the workload and flower for the person of putting, and improves the efficiency and accuracy of decision, it is efficient, convenient, accurate to meet advertisement putting person
The advertisement that makes a choice shows the demand of the decision of the page.
Further, for being believed according to the history page clicking rate information and the apriori probability Distribution Value that obtain each page
Breath, the submodule for calculating the posterior probability distributed intelligence of page-out clicking rate is the bayesian reasoning for carrying out operation according to the following equation
Computing module:
Wherein, θ is that Probability Distribution parameter (distribution parameter), p (θ) are apriori probability distribution in advance
(prior distribution), p (x | θ ') it is similar function (likelihood function), p (θ | x) posterior probability point
Cloth (posterior distribution).It will be understood to those skilled in the art that with bayesian reasoning (Bayesian
Inference the posterior probability distributed intelligence for) calculating each page clicking rate reduces advertisement putting person to this numerical value of click and highest
The advertisement of clicking rate shows the artificial statistics and artificial judgment of the page, reduces cost.
Further, described to record for being clicked according to history, the submodule for calculating the display number accounting of each page is, when
The cumulative number accounting value that the page that number is i is shown is bi, history page that number is j shows that the sample size of information (is adopted
Sample number) it is njWhen, the computing module that is calculated in the following way:
Further, when the probability density function of the posterior probability distributed intelligence is fj(x), the posterior probability distribution letter
The probability cumulative density function of breath is Fj(x) when, relationship between the two is
Further, the probability density function of the posterior probability distributed intelligence for the page for being i when number
(probability density function, p.d.f.) is fi(x), the posterior probability point for the page for being j when number
The probability cumulative density function (cumulated density function, c.d.f.) of cloth information is Fj(x) and number is i
The page the winning rate piBetween relationship be
Further, the recommendation page number is k=argmax (pi-bi).In other words, the winning rate p of recommendation page kkWith
The accumulative display number accounting b of the recommendation pagekThe difference p of the twok-bkGreater than the winning rate p of other pages iiWith the page
Accumulative display number accounting biThe difference p of the twoi-bi。
Further, the module for exporting recommendation page results information includes:
For showing the submodule for recommending page results；
User selects the corresponding display page according to the recommendation page results of display and generates new history page clicking rate
The submodule of information；
It is used to be clicked according to the history page for obtaining each page for feeding back to new history page clicking rate information
Rate information and the apriori probability are distributed value information, calculate the submodule of the posterior probability distributed intelligence of page-out clicking rate.
It will be understood to those skilled in the art that being fed back by new history page clicking rate information, the present embodiment is mentioned
The page clicking rate data processing equipment of confession can form self-study mechanism, accumulate new history page clicking rate information, this field
It is latent that technical staff is appreciated that the huger recommendation page results to acquisition of the data of history page clicking rate information more have
Power.
Embodiment three
As shown in figure 3, page clicking rate data system provided by the present embodiment, including client, the client packet
It includes: microprocessor controller, display screen and touch screen；The display screen and touch screen connect with the microprocessor controller circuit
It connects.It will be understood by those skilled in the art that the display screen is for showing advertisement page；The touch screen is used for when user is to this
The page is clicked by touch screen when page display content is interested, obtains the page info to realize；The micro process control
Device processed is used for the history page point of the apriori probability Distribution Value and the page got of the page clicking rate of the setting according to user
Maximum clicking rate result is calculated to helping user to learn which page clicking rate is most in the rate information of hitting, user, such as advertisement
Putting person can determine that selecting the dispensing advertisement of that page that can bring more preferably publicizes by learning which page clicking rate at most
Effect.
As shown in figure 4, page clicking rate data system provided by the present embodiment, further include cloud data server 1 and
Communication network 2, a plurality of microprocessor controllers realize that data are logical by communication network 2 and the cloud data server 1
News connection.In this way, the data of acquisition can be uploaded to cloud data server 1 by microprocessor controller runs operation, so as to obtain
Huge data system is obtained, operation result and decision conclusions are more accurate and stable.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (23)
1. a kind of page clicking rate data processing method, which comprises the following steps:
S1 is used to obtain the step of page clicking rate information of the different pages respectively；
S2, which is used to be analyzed and processed the page clicking rate information of the different pages of acquisition and generate, recommends page results information
The step of；
S3 is for exporting the step of recommending page results information.
2. the method as described in claim 1, which is characterized in that the S1 is used to obtain the page clicking rate of the different pages respectively
The step of information includes:
S11 sets the apriori probability distribution value information of each page clicking rate；
S12 obtains the history page clicking rate information of each page.
3. method according to claim 2, which is characterized in that when the Probability Distribution mould in advance for the page clicking rate that number is i
Type is in Gaussian Profile (Gaussian distribution), and Probability Distribution parameter is average value mu in advanceiAnd standard deviation sigmai(>0)
When, probability density function is
4. method as claimed in claim 3, which is characterized in that the S2 is used for the page clicking rate to the different pages of acquisition
Information is analyzed and processed and includes: the step of generating-recommend page results information
S21 is distributed value information according to the history page clicking rate information and the apriori probability that obtain each page, calculates page
The posterior probability distributed intelligence of face clicking rate；
S22 is clicked according to history and is recorded, and calculates the display number accounting of each page；
S23 calculates the winning rate of each page according to the posterior probability distributed intelligence；It will be understood by those skilled in the art that described
The winning rate of the page refers to that the page becomes the percentage of winning for recommending the page；
S24 calculates the display number accounting and winning rate posterior probability distribution letter of each page obtained according to above-mentioned steps
Breath, which calculates, recommends page results information.
5. method as claimed in claim 4, which is characterized in that described to be believed according to the history page clicking rate for obtaining each page
The step of breath and the apriori probability are distributed value information, calculate the posterior probability distributed intelligence of page-out clicking rate pushes away with bayesian
The method of reason is calculated, to obtain the posterior probability distributed intelligence of page-out clicking rate.
6. method as claimed in claim 5, which is characterized in that the method that the method with bayesian reasoning is calculated
Are as follows:
Wherein, θ is Probability Distribution parameter in advance, and p (θ) is that apriori probability is distributed, and p (x | θ ') is similar function, and p (θ | x) is thing
Probability distribution afterwards.
7. method as claimed in claim 6, which is characterized in that the S22 is clicked according to history and recorded, and calculates the aobvious of each page
In the step of showing number accounting, when the cumulative number accounting value that the page that number is i is shown is bi, history page that number is j
The sample size (sampling number) for showing information is njWhen, it is calculated in the following way:
8. the method for claim 7, which is characterized in that when the probability density function of the posterior probability distributed intelligence is
fj(x), the probability cumulative density function of the posterior probability distributed intelligence is Fj(x) when, relationship between the two is
9. method according to claim 8, which is characterized in that when the posterior probability distributed intelligence for the page that number is i
Probability density function be fi(x), the probability cumulative density function of the posterior probability distributed intelligence for the page for being j when number
For Fj(x) and number be i the page the winning rate piBetween relationship be
10. method as claimed in claim 9, which is characterized in that the winning rate p for recommending page kkWith tiring out for the recommendation page
The display number accounting b of meterkThe difference p of the twok-bkGreater than the winning rate p of other pages iiWith the accumulative display number of the page
Accounting biThe difference p of the twoi-bi。
11. the method as described in any one of claim 2 to 10, which is characterized in that the S3 is for exporting maximum clicking rate
The step of result information includes:
Page results are recommended in S31 display；
S32 user selects the corresponding display page according to the recommendation page results of display and generates new history page clicking rate letter
Breath；
S33 return step S12.
12. a kind of page clicking rate data processing equipment characterized by comprising
For obtaining the module of the page clicking rate information of the different pages respectively；
Page clicking rate information for the different pages to acquisition, which is analyzed and processed and generates, recommends page results information
Module；
For exporting the module for recommending page results information.
13. page clicking rate data processing equipment as claimed in claim 12, which is characterized in that described for obtaining respectively not
Module with the page clicking rate information of the page includes:
Apriori probability for setting each page clicking rate is distributed the submodule of value information；
For obtaining the submodule of the history page clicking rate information of each page.
14. page clicking rate data processing equipment as claimed in claim 13, which is characterized in that when number be i page point
The model of Probability Distribution in advance of rate is hit in Gaussian Profile (Gaussian distribution), Probability Distribution parameter is flat in advance
Mean μiAnd standard deviation sigmaiWhen (> 0), probability density function is
15. page clicking rate data processing equipment as claimed in claim 14, which is characterized in that described to be used for acquisition not
Page clicking rate information with the page be analyzed and processed and generate recommend page results information module include:
For being distributed value information according to the history page clicking rate information and the apriori probability that obtain each page, page is calculated
The submodule of the posterior probability distributed intelligence of face clicking rate；
It is recorded for being clicked according to history, calculates the submodule of the display number accounting of each page；
For calculating the winning rate of each page according to the posterior probability distributed intelligence；
The display number accounting and the winning rate posterior probability for calculating each page obtained according to above-mentioned steps are distributed
Information calculates the submodule for recommending page results information.
16. page clicking rate data processing equipment as claimed in claim 15, which is characterized in that for according to each page of acquisition
The history page clicking rate information in face and the apriori probability are distributed value information, calculate the posterior probability distribution of page-out clicking rate
The submodule of information is the bayesian reasoning computing module for carrying out operation according to the following equation:
Wherein, θ is that Probability Distribution parameter (distribution parameter), p (θ) they are that apriori probability is distributed in advance, p (x |
θ ') it is similar function, the distribution of p (θ | x) posterior probability.
17. the page clicking rate data processing equipment as described in claim 15 or 16, which is characterized in that described to be gone through for basis
History clicks record, and the submodule for calculating the display number accounting of each page is, when the cumulative number that the page that number is i is shown accounts for
Ratio is bi, the sample size for the history page display information that number is j is njWhen, the operation that is calculated in the following way
Module:
18. page clicking rate data processing equipment as claimed in claim 17, which is characterized in that when the posterior probability is distributed
The probability density function of information is fj(x), the probability cumulative density function of the posterior probability distributed intelligence is Fj(x) when, the two
Between relationship be
19. page clicking rate data processing equipment as claimed in claim 18, which is characterized in that when number be i the page
The probability density function of the posterior probability distributed intelligence is fi(x), the posterior probability for the page for being j when number is distributed letter
The probability cumulative density function of breath is Fj(x) and number be i the page the winning rate piBetween relationship be
20. page clicking rate data processing equipment as claimed in claim 19, which is characterized in that the recommendation page number isIn other words, the winning rate p of recommendation page kkWith the accumulative display number accounting b of the recommendation pagek
The difference p of the twok-bkGreater than the winning rate p of other pages iiWith the accumulative display number accounting b of the pageiThe difference p of the twoi-
bi。
21. the page clicking rate data processing equipment as described in any one of claim 12 to 16, which is characterized in that the use
Include: in the module for exporting maximum clicking rate result information
For showing the submodule for recommending page results；
User selects the corresponding display page according to the recommendation page results of display and generates new history page clicking rate information
Submodule；
For new history page clicking rate information to be fed back to the history page clicking rate letter being used for according to each page is obtained
Breath and the apriori probability are distributed value information, calculate the submodule of the posterior probability distributed intelligence of page-out clicking rate.
22. a kind of page clicking rate data system, which is characterized in that including client, the client includes: micro process control
Device, display screen and touch screen；The display screen and touch screen with the microprocessor controller circuit connection.
23. page clicking rate data system as claimed in claim 22, which is characterized in that further include cloud data server
(1) and communication network (2), a plurality of microprocessor controllers pass through communication network (2) and the cloud data server (1)
Realize data communication connection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710542895.5A CN109214847A (en) | 2017-07-05 | 2017-07-05 | Page clicking rate data processing method, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710542895.5A CN109214847A (en) | 2017-07-05 | 2017-07-05 | Page clicking rate data processing method, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109214847A true CN109214847A (en) | 2019-01-15 |
Family
ID=64993670
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710542895.5A Pending CN109214847A (en) | 2017-07-05 | 2017-07-05 | Page clicking rate data processing method, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109214847A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111160953A (en) * | 2019-12-12 | 2020-05-15 | 浙江大华技术股份有限公司 | Method and device for delivering advertisement in application and storage medium |
| CN112633930A (en) * | 2020-12-25 | 2021-04-09 | 上海硬通网络科技有限公司 | Page resource delivery method and device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101866341A (en) * | 2009-04-17 | 2010-10-20 | 华为技术有限公司 | Information push method, device and system |
| CN103177371A (en) * | 2011-12-21 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Method and device for exhibiting information |
| CN103942603A (en) * | 2013-01-17 | 2014-07-23 | 腾讯科技（深圳）有限公司 | Advertisement click rate prediction method and device |
| CN105117491A (en) * | 2015-09-22 | 2015-12-02 | 北京百度网讯科技有限公司 | Page pushing method and device |
-
2017
- 2017-07-05 CN CN201710542895.5A patent/CN109214847A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101866341A (en) * | 2009-04-17 | 2010-10-20 | 华为技术有限公司 | Information push method, device and system |
| CN103177371A (en) * | 2011-12-21 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Method and device for exhibiting information |
| CN103942603A (en) * | 2013-01-17 | 2014-07-23 | 腾讯科技（深圳）有限公司 | Advertisement click rate prediction method and device |
| CN105117491A (en) * | 2015-09-22 | 2015-12-02 | 北京百度网讯科技有限公司 | Page pushing method and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111160953A (en) * | 2019-12-12 | 2020-05-15 | 浙江大华技术股份有限公司 | Method and device for delivering advertisement in application and storage medium |
| CN111160953B (en) * | 2019-12-12 | 2023-11-03 | 浙江大华技术股份有限公司 | Method, device and storage medium for putting advertisement in application |
| CN112633930A (en) * | 2020-12-25 | 2021-04-09 | 上海硬通网络科技有限公司 | Page resource delivery method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104992347B (en) | A kind of method and device of video matching advertisement | |
| CN104462245B (en) | A kind of user's online preference data recognition methods | |
| CN109359521A (en) | The two-way assessment system of Classroom instruction quality based on deep learning | |
| WO2018132840A1 (en) | Methods of assessing long-term indicators of sentiment | |
| CN107038480A (en) | A kind of text sentiment classification method based on convolutional neural networks | |
| CN107396144B (en) | A kind of barrage distribution method and device | |
| CN109299344A (en) | The generation method of order models, the sort method of search result, device and equipment | |
| CN107483982A (en) | A kind of main broadcaster recommends method and apparatus | |
| CN107798027B (en) | Information popularity prediction method, information recommendation method and device | |
| CN109905696A (en) | A kind of recognition methods of the Video service Quality of experience based on encryption data on flows | |
| US11269963B1 (en) | Estimating social content interactions | |
| CN109547546A (en) | A kind of dispatching method and control centre's server of request task | |
| CN108009220A (en) | A kind of method for being detected in network hotspot public sentiment event and positioning abnormal user | |
| CN107609147A (en) | A kind of method and system that feature is automatically extracted from log stream | |
| CN104348747A (en) | Method and system for monitoring flow in MPLS-VPN (Multiple Protocol Label Switching-Virtual Private Network) | |
| WO2023125460A1 (en) | Sorting model training method and sorting method | |
| CN109214847A (en) | Page clicking rate data processing method, apparatus and system | |
| CN104935563B (en) | Method and apparatus for managing meeting | |
| CN106127519A (en) | A kind of live platform user divided method based on K Means algorithm and system | |
| CN108710635B (en) | Content recommendation method and device | |
| WO2021233015A1 (en) | Information processing method and apparatus, and computer readable storage medium | |
| CN104731788B (en) | The processing method and equipment of promotion message | |
| CN105184646B (en) | Self-help bank's information-pushing method and device | |
| CN110457387B (en) | Method and related device applied to user tag determination in network | |
| CN108351903A (en) | Use the optimization of the multi-channel system of feedback control loop |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190115 |
|
| WD01 | Invention patent application deemed withdrawn after publication |