CN106600342A - Advertisement delivery method and device - Google Patents
Advertisement delivery method and device Download PDFInfo
- Publication number
- CN106600342A CN106600342A CN201611248934.2A CN201611248934A CN106600342A CN 106600342 A CN106600342 A CN 106600342A CN 201611248934 A CN201611248934 A CN 201611248934A CN 106600342 A CN106600342 A CN 106600342A
- Authority
- CN
- China
- Prior art keywords
- user
- classification
- fraction
- degree
- under
- 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
- G06Q30/0251—Targeted advertisements
- G06Q30/0257—User requested
-
- 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
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an advertisement delivery method which comprises a step of obtaining a seed user in advertisement historical data, wherein the seed user is a user which completely watches the remaining part of an advertisement, a step of calculating the scores of multiple categories of video users through a similarity algorithm with interest, novelty and quality as parameters, and selecting a target user from the users in multiple categories according to the scores of the multiple categories, and a step of carrying out advertisement delivery on the target user. Thus the scores of the categories are calculated with a mode of the combination of the interest, novelty and quality, the calculated scores of the categories are more accurate, the target user is selected according to the more accurate score of a category, thus the target user is more accurate, and the fast and accurate determination of the target user to carry out advertisement delivery is realized. In addition, the invention also discloses an advertisement delivery device.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of method and apparatus of advertisement putting.
Background technology
At present, increasing enterprise-like corporation selects to throw in advertisement in video to publicize corporate culture, product etc., passes through
With the interaction of video user, power-assisted brand, marketing objectives is realized.It is now with a kind of new advertisement broadcast format, i.e., a kind of preposition
Advertisement, when advertising unit is commenced play out, it may appear that countdown for 5 seconds, after having countdowned, user can see and skip
The button of the remainder of advertisement, clicks on button and skips advertisement, then proceedes to browse the content for wishing viewing, or, Yong Huye
Button can not be clicked on and select the remainder for continuing to watch advertisement.Part advertiser prefers this advertising format, because
It is to broadcast flow just for the advertisement that " skipping " button is not clicked in user to be paid, it means that at least in theory,
Without click on " skipping " button user be really to watch advertisement it is interested.
When advertiser selects above-mentioned advertisement broadcast format, in the prior art, by not having for the publicity orders in history
The user of remainder of advertisement is skipped as seed user, is counted as major parameter with the ratio of the seed user under multiple labels
The fraction of the multiple labels of video user is calculated, is sorted from big to small according to the fraction of each label, choose whole successively according to sequence
User under label, until when number of users exceedes advertising presupposition input crowd's amount size, as targeted customer line upslide is entered
Put.
Inventor has found that the ratio in prior art with the seed user under multiple labels is counted as major parameter through research
The fraction of the multiple labels of video user is calculated, due to, compared with the degree of overlapping for focusing on being user and seed user under label, causing when calculating
The fraction of the label for calculating is not accurate enough, so as to cause the targeted customer of last selection not accurate enough.
Based on this, in this day and age, how more accurately to determine that targeted customer carries out advertisement putting and becomes what urgent need was solved
Technical problem.
The content of the invention
The technical problem to be solved is to provide a kind of method and apparatus of advertisement putting, with interest-degree, novelty
Degree and the mode of quality triplicity remove the fraction for calculating classification, and according to the fraction of more accurately classification targeted customer is chosen, and enter
And more accurately determine that targeted customer carries out advertisement putting.
To solve above-mentioned technical problem, the invention provides a kind of method of advertisement putting, the method includes:
The seed user in history of advertising data is obtained, the seed user is the user for completely watching the advertisement;
With interest-degree, novelty degree and quality as parameter, by dividing for multiple classifications of Similarity Algorithm calculating video user
Number, the interest-degree refers to the ratio of seed user under the classification, and the novel degree refers to other users under the classification
Ratio, the quality refers to the ratio of seed user under classification described in preset number of days historical data, and the other users are class
The user not descended beyond seed user;
According to the fraction of the plurality of classification, targeted customer is chosen in user under the plurality of classification;
Advertisement putting is carried out to the targeted customer.
Preferably, the formula of the Similarity Algorithm calculating is:
Score=a × ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction,
Log is natural logrithm, and the interest is interest-degree, and the novelty is novelty degree, and the quality is quality, described
A, b, c are predetermined coefficient.
Preferably, the interest-degree be equal under the seed user and the classification common factor number of users of user with it is described
The ratio of number of users under classification, the novel degree subtracts the interest-degree equal to 1, and the quality is equal to preset number of days historical data
Described under classification seed user quantity and number of users under classification described in preset number of days historical data ratio.
Preferably, the fraction according to the plurality of classification, targeted customer is chosen under the plurality of classification in user,
Including:
Using the best result in the fraction of all categories belonging to each user as user fraction, according to each user's
Fraction order from big to small is ranked up to each user;
Targeted customer is chosen in each user sorted under the plurality of classification.
Preferably, the fraction according to the plurality of classification, targeted customer is chosen under the plurality of classification in user,
Specially:Crowd's amount is thrown according to the fraction and advertising presupposition of the plurality of classification, is chosen from user under the plurality of classification
Targeted customer.
Present invention also offers a kind of device of advertisement putting, the device includes:
Acquiring unit, for obtaining history of advertising data in seed user, the seed user is described for complete viewing
The user of advertisement;
Computing unit, for interest-degree, novelty degree and quality as parameter, by Similarity Algorithm video user being calculated
The fraction of multiple classifications, the interest-degree refers to the ratio of seed user under the classification, and the novel degree refers to the classification
The ratio of lower other users, the quality refers to the ratio of seed user under classification described in preset number of days historical data, described
Other users are the user under classification beyond seed user;
First chooses unit, for according to the fraction of the plurality of classification, mesh being chosen in user under the plurality of classification
Mark user;
Unit is thrown in, for carrying out advertisement putting to the targeted customer.
Preferably, the formula of the Similarity Algorithm calculating is:
Score=a × ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction,
Ln is natural logrithm, and the interest is interest-degree, and the novelty is novelty degree, and the quality is quality, described
A, b, c are predetermined coefficient.
Preferably, the interest-degree be equal under the seed user and the classification common factor number of users of user with it is described
The ratio of number of users under classification, the novel degree subtracts the interest-degree equal to 1, and the quality is equal to preset number of days historical data
Described under classification seed user quantity and number of users under classification described in preset number of days historical data ratio.
Preferably, the first selection unit includes:
Sequencing unit, for using the best result in the fraction of all categories belonging to each user as user fraction,
Fraction order from big to small according to each user is ranked up to each user;
Second chooses unit, for choosing targeted customer in each user of sequence under the plurality of classification.
Preferably, described first unit is chosen specifically for the fraction according to the plurality of classification and advertising presupposition input people
Group's amount, targeted customer is chosen under the plurality of classification in user.
Compared with prior art, the present invention has advantages below:
In embodiments of the present invention, the seed user in history of advertising data is obtained, the seed user is complete viewing
The user of the advertisement;With interest-degree, novelty degree and quality as parameter, multiple classes of video user are calculated by Similarity Algorithm
Other fraction;According to the fraction of the plurality of classification, targeted customer is chosen in user under the plurality of classification;To the target
User carries out advertisement putting.As can be seen here, go to calculate dividing for classification in the way of interest-degree, novelty degree and quality triplicity
Number so that the fraction of the classification for calculating is more accurate, according to the fraction of more accurately classification targeted customer is chosen so that target is used
Family is also more accurate, it is achieved thereby that accurately determining that targeted customer carries out advertisement putting.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present application or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments described in application, for those of ordinary skill in the art, on the premise of not paying creative work,
Can be with according to these other accompanying drawings of accompanying drawings acquisition.
Fig. 1 is the block schematic illustration of an exemplary application scene in the embodiment of the present invention;
Fig. 2 is a kind of schematic flow sheet of the method for advertisement putting in the embodiment of the present invention;
Fig. 3 is the schematic flow sheet of the method for another kind of advertisement putting in the embodiment of the present invention;
Fig. 4 is a kind of structural representation of the device of advertisement putting in the embodiment of the present invention;
Fig. 5 is the structural representation of the device of another kind of advertisement putting in the embodiment of the present invention.
Specific embodiment
In order that those skilled in the art more fully understand application scheme, below in conjunction with the embodiment of the present application
Accompanying drawing, is clearly and completely described to the technical scheme in the embodiment of the present application, it is clear that described embodiment is only this
Apply for a part of embodiment, rather than the embodiment of whole.Based on the embodiment in the application, those of ordinary skill in the art exist
The every other embodiment obtained under the premise of creative work is not made, the scope of the application protection is belonged to.
Inventor has found that the ratio in prior art with the seed user under multiple labels is counted as major parameter through research
The fraction of the multiple labels of video user is calculated, due to, compared with the degree of overlapping for focusing on being user and seed user under label, causing when calculating
The fraction of the label for calculating is not accurate enough, and the user chosen successively according to the sequence of label under whole label adds seed user
As targeted customer, only have certain customers and seed user similarity high in whole label in fact, therefore choose under whole label
User cause targeted customer not accurate enough, when advertising presupposition throw in crowd amount it is larger when, need to choose the use under multiple labels
Family, chooses process and takes seriously, and the situation that the user under multiple labels very likely duplicates also needs to sieve again after selection
Choosing, process is complicated.
Based on this, in embodiments of the present invention, the seed user in history of advertising data is obtained, the seed user has been
The whole user for watching the advertisement;With interest-degree, novelty degree and quality as parameter, video user is calculated by Similarity Algorithm
The fraction of multiple classifications;According to the fraction of the plurality of classification, targeted customer is chosen in user under the plurality of classification;To institute
Stating targeted customer carries out advertisement putting.As can be seen here, go to calculate classification in the way of interest-degree, novelty degree and quality triplicity
Fraction so that the fraction of the classification for calculating is more accurate, is not to choose the user under whole label successively, but first by each
Best result in the fraction of all categories belonging to user as user fraction, according to each user fraction from big to small
Order is ranked up to each user, throws in crowd's amount further according to advertising presupposition and directly chooses from each user of the sequence
Targeted customer, the targeted customer of selection is more accurate, and selection process take it is short, it is achieved thereby that fast accurate ground determine target
User carries out advertisement putting.
For example, one of scene of the embodiment of the present invention, can be applied to scene as shown in Figure 1.At this
Jing Zhongyou has in this scenario server 101 and client 102, and server 101 and client 102 can be interacted, server 101
The seed user in history of advertising data is obtained, the seed user is the user for completely watching the advertisement;Server 101 with
Interest-degree, novelty degree and quality are parameter, and the fraction of multiple classifications of video user, the interest are calculated by Similarity Algorithm
Degree refers to the ratio of seed user under the classification, and the novel degree refers to the ratio of other users under the classification, the matter
Amount refers to the ratio of seed user under classification described in preset number of days historical data, and the other users are seed user under classification
Other users in addition;Server 101 chooses target under the plurality of classification according to the fraction of the plurality of classification in user
User;Server 101 is according to the targeted customer by advertisement putting in client 102.
It is understood that in above-mentioned application scenarios, although by the action description of embodiment of the present invention by server
101 perform, but the present invention is unrestricted in terms of executive agent, as long as performing the action disclosed in embodiment of the present invention
.
It is understood that above-mentioned scene is only a Sample Scenario provided in an embodiment of the present invention, the embodiment of the present invention
It is not limited to this scene.
Below in conjunction with the accompanying drawings, by embodiment to describe the embodiment of the present invention in detail in advertisement putting method and apparatus
Specific implementation.
Illustrative methods
Referring to Fig. 2, a kind of schematic flow sheet of the method for advertisement putting in the embodiment of the present invention is shown.In the present embodiment
In, methods described for example may comprise steps of:
Step 201:The seed user in history of advertising data is obtained, the seed user is completely to watch the advertisement
User.
Step 202:With interest-degree, novelty degree and quality as parameter, the multiple of video user are calculated by Similarity Algorithm
The fraction of classification, the interest-degree refers to the ratio of seed user under the classification, and the novel degree refers under the classification it
The ratio of his user, the quality refers to the ratio of seed user under classification described in preset number of days historical data, it is described other
User is the user under classification beyond seed user.
It is understood that the interest-degree is equal to the common factor number of users of user under the seed user and the classification
With the ratio of number of users under the classification, the novel degree subtracts the interest-degree equal to 1, and the quality is gone through equal to preset number of days
Described in history data under classification described in seed user quantity and preset number of days historical data under classification number of users ratio.Example
Such as, the total 50w of seed user, the total 1kw of user under male's classification, it is 25w that both occur simultaneously, in nearest five days that non-directional is thrown in
The total 20w of user tag crowd under male's classification, the total 5w of seed user under male's classification in nearest five days, then
Interest-degree is equal to the ratio of 25w and 1kw, and novel degree is equal to 1 ratio for subtracting the 25w and 1kw, and quality is equal to the ratio of 5w and 20w
Value.
In some embodiments of the present embodiment, the formula that the Similarity Algorithm is calculated for example is specifically as follows:
Score=a × ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction, and log is certainly
Right logarithm, the interest is interest-degree, and the novelty is novelty degree, and the quality is quality, and a, b, c are
Predetermined coefficient.Go to calculate dividing for classification in the way of different specific weight combines using interest-degree, three parameters of novelty degree and quality
Number, the fraction for calculating is more accurate.For example, as predetermined coefficient a, b, when c is respectively 0.6,0.0001,0.4, then similarity
Algorithm calculate formula be:Score=0.6 × ln (interest)+0.00001 × ln (novelty)+0.4 × ln
(quality), the wherein interest-degree and quality proportion in fraction is larger.
Step 203:According to the fraction of the plurality of classification, targeted customer is chosen in user under the plurality of classification.
It should be noted that the fraction of the plurality of classification has been obtained by step 202, if the directly fraction to classification
Sort from big to small, choose the user under whole classification successively by sequence, it may appear that although the selection greatly of classification fraction is whole
Need when user repeats in situation that the certain customers of the user under classification are not interested in advertisement or multiple classifications of selection
The situation of user is screened, in order to ensure the targeted customer for choosing is more accurate, can be adopted to each user's sequence, then directly selected
The user higher with seed user similarity is selected as targeted customer.In some embodiments of the present embodiment, step 203
Such as can include:Using the best result in the fraction of all categories belonging to each user as the fraction of user, use according to each
The fraction at family order from big to small is ranked up to each user;Choose in each user sorted under the plurality of classification
Targeted customer.For example, video user has male, and 20-30 year, student, five classifications of women and worker, their fraction is respectively
70,60,80,50 and 90, when all categories belonging to user a are male, 20-30 year and during student, then the fraction of user a is
80, when all categories belonging to user b are male, 20-30 year and during worker, then the fraction of user b is 90, when belonging to user c
All categories be 20-30 year, when student and women, then the fraction of user c is 80, therefore user a, b and c are ordered as b>a
=c, i.e., before b comes, behind a and c is neck and neck.
It should be noted that generally publicity orders include that advertising presupposition throws in crowd's amount, it is disposable in order to meet
The quantity of the targeted customer of selection throws in crowd's amount more than advertising presupposition, can throw in crowd's amount according to advertising presupposition and go selection one
The targeted customer of fixed number amount.In some embodiments of the present embodiment, step 203 is for example specifically as follows:According to described many
The fraction of individual classification and advertising presupposition throw in crowd's amount, and targeted customer is chosen in user under the plurality of classification.For example, advertisement
Default input crowd amount is 100w, and 100w user is targeted customer before choosing from the user of sequence.
Step 204:Advertisement putting is carried out to the targeted customer.
It should be noted that advertisement putting is carried out to the targeted customer, according to the data thrown in after terminating, it can be found that
The ratio of seed user is greatly improved in targeted customer.
The various embodiments provided by the present embodiment, obtain the seed user in history of advertising data, the seed
User is the user for completely watching the advertisement;With interest-degree, novelty degree and quality as parameter, calculated by Similarity Algorithm and regarded
The fraction of multiple classifications of frequency user;According to the fraction of the plurality of classification, target is chosen in user under the plurality of classification
User;Advertisement putting is carried out to the targeted customer.As can be seen here, in the way of interest-degree, novelty degree and quality triplicity
Remove the fraction for calculating classification so that the fraction of the classification for calculating is more accurate, is not to choose the user under whole label successively, and
Be first using the best result in the fraction of all categories belonging to each user as user fraction, according to the fraction of each user
Order from big to small is ranked up to each user, further according to advertising presupposition throw in crowd amount directly from the sequence each
Targeted customer is chosen in user, the targeted customer of selection is more accurate, and selection process take it is short, it is achieved thereby that fast accurate
Ground determines that targeted customer carries out advertisement putting.
Referring to Fig. 3, the schematic flow sheet of the method for another kind of advertisement putting in the embodiment of the present invention is shown.In this enforcement
In example, methods described for example may comprise steps of:
Step 301:According to publicity orders, non-directional input is carried out to the advertisement, the publicity orders include default throwing
Put crowd's amount.
Step 302:The seed user in the history of advertising data is obtained, the seed user is described wide completely to watch
The user of announcement.
Step 303:Whether the seed user quantity is judged more than threshold value, the threshold value is according to the default input people
Group's amount is arranged, if it is, into step 304;If not, return to step 301.
For example, when the default input crowd amount of publicity orders is 100w, threshold value can be set for 25w.
Step 304:With interest, novelty, quality as parameter, using Similarity Algorithm video user is calculated
The fraction of multiple classifications, the interest be under the seed user and the classification common factor number of users of user with it is described
The ratio of number of users under classification, the novelty subtracts the interest-degree for 1, and the quality is preset number of days historical data
Described under classification seed user quantity and number of users under classification described in preset number of days historical data ratio.
It should be noted that more focusing on the ratio of seed user and preset number of days historical data under the classification when calculating
Described under classification seed user ratio, so that the fraction for calculating is more accurate, i.e., user uses with seed under described classification
The similarity at family is higher;The ratio for increasing other users under the classification during calculating is under the expandtabs in targeted customer
Other users, the other users are the user under classification beyond seed user.In some embodiments of the present embodiment, institute
The computing formula for stating Similarity Algorithm is for example specifically as follows:Score=0.6 × ln (interest)+0.00001 × ln
(novelty)+0.4×ln(quality)。
Step 305:Using the best result in the fraction of all categories belonging to each user as user fraction, according to each
The fraction of individual user order from big to small is ranked up to each user.
Step 306:Targeted customer is chosen from each user of the sequence according to the default input crowd amount is carried out
Advertisement putting.
The various embodiments provided by the present embodiment, obtain the seed user in history of advertising data, the seed
User is the user for completely watching the advertisement;With interest-degree, novelty degree and quality as parameter, calculated by Similarity Algorithm and regarded
The fraction of multiple classifications of frequency user;According to the fraction of the plurality of classification, target is chosen in user under the plurality of classification
User;Advertisement putting is carried out to the targeted customer.As can be seen here, in the way of interest-degree, novelty degree and quality triplicity
Remove the fraction for calculating classification so that the fraction of the classification for calculating is more accurate, is not to choose the user under whole label successively, and
Be first using the best result in the fraction of all categories belonging to each user as user fraction, according to the fraction of each user
Order from big to small is ranked up to each user, further according to advertising presupposition throw in crowd amount directly from the sequence each
Targeted customer is chosen in user, the targeted customer of selection is more accurate, and selection process take it is short, it is achieved thereby that fast accurate
Ground determines that targeted customer carries out advertisement putting.
Example devices
Referring to Fig. 4, a kind of structural representation of the device of advertisement putting in the embodiment of the present invention is shown.In the present embodiment
In, described device for example can include:
Acquiring unit 401, for obtaining history of advertising data in seed user, the seed user is complete viewing institute
State the user of advertisement.
Computing unit 402, for interest-degree, novelty degree and quality as parameter, calculating video by Similarity Algorithm and using
The fraction of multiple classifications at family, the interest-degree refers to the ratio of seed user under the classification, and the novel degree refers to described
The ratio of other users under classification, the quality refers to the ratio of seed user under classification described in preset number of days historical data,
The other users are the other users under classification beyond seed user.
Optionally, in some embodiments of the present embodiment, the formula that the Similarity Algorithm is calculated is:Score=a
× ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction, and ln is natural logrithm, institute
It is interest-degree to state interest, and the novelty is novelty degree, and the quality is quality, and a, b, c are predetermined coefficient.
Optionally, in some embodiments of the present embodiment, the interest-degree is equal to the seed user and the class
The common factor number of users of user and the ratio of number of users under the classification, the novel degree is not descended to subtract the interest-degree equal to 1,
The quality is equal to described in seed user quantity under classification described in preset number of days historical data and preset number of days historical data
The ratio of number of users under classification.
First chooses unit 403, for according to the fraction of the plurality of classification, choosing from user under the plurality of classification
Targeted customer.
Optionally, in some embodiments of the present embodiment, the first selection unit 403 for example can include:
Sequencing unit, for using the best result in the fraction of all categories belonging to each user as user fraction,
Fraction order from big to small according to each user is ranked up to each user;
Second chooses unit, for choosing targeted customer in each user of sequence under the plurality of classification.
Optionally, in some embodiments of the present embodiment, the first selection unit 403 for example specifically can be used for
Crowd's amount is thrown according to the fraction and advertising presupposition of the plurality of classification, target is chosen from user under the plurality of classification and is used
Family.
Unit 404 is thrown in, for carrying out advertisement putting to the targeted customer.
The various embodiments provided by the present embodiment, obtain the seed user in history of advertising data, the seed
User is the user for completely watching the advertisement;With interest-degree, novelty degree and quality as parameter, calculated by Similarity Algorithm and regarded
The fraction of multiple classifications of frequency user;According to the fraction of the plurality of classification, target is chosen in user under the plurality of classification
User;Advertisement putting is carried out to the targeted customer.As can be seen here, in the way of interest-degree, novelty degree and quality triplicity
Remove the fraction for calculating classification so that the fraction of the classification for calculating is more accurate, is not to choose the user under whole label successively, and
Be first using the best result in the fraction of all categories belonging to each user as user fraction, according to the fraction of each user
Order from big to small is ranked up to each user, further according to advertising presupposition throw in crowd amount directly from the sequence each
Targeted customer is chosen in user, the targeted customer of selection is more accurate, and selection process take it is short, it is achieved thereby that fast accurate
Ground determines that targeted customer carries out advertisement putting.
Referring to Fig. 5, the structural representation of the device of another kind of advertisement putting in the embodiment of the present invention is shown.In this enforcement
In example, described device for example can include:
First throws in unit 501, and for according to publicity orders, non-directional input being carried out to the advertisement, the advertisement is ordered
It is single to include default input crowd amount.
Acquiring unit 502, for obtaining the history of advertising data in seed user, the seed user is complete sight
See the user of the advertisement.
Judging unit 503, for whether judging the seed user quantity more than threshold value, if it is, into step 304;
If not, return to step 301, the threshold value is arranged according to the default input crowd amount.
Computing unit 504, for interest, novelty, quality as parameter, being calculated using Similarity Algorithm and being regarded
The fraction of multiple classifications of frequency user, the interest is the common factor number of users of user under the seed user and the classification
The ratio with number of users under the classification is measured, the novelty subtracts the interest-degree for 1, and the quality is preset number of days
Described in historical data under classification described in seed user quantity and preset number of days historical data under classification number of users ratio.
Optionally, in some embodiments of the present embodiment, the computing formula of the Similarity Algorithm for example specifically may be used
Think:Score=0.6 × ln (interest)+0.00001 × ln (novelty)+0.4 × ln (quality).
Sequencing unit 505, for the dividing as user using the best result in the fraction of all categories belonging to each user
Number, is ranked up according to the fraction order from big to small of each user to each user.
Second throws in unit 506, for being chosen from each user of the sequence according to the default input crowd amount
Targeted customer carries out advertisement putting.
The various embodiments provided by the present embodiment, obtain the seed user in history of advertising data, the seed
User is the user of the remainder for completely watching the advertisement;With interest-degree, novelty degree and quality as parameter, by similarity
Algorithm calculates the fraction of multiple classifications of video user;According to the fraction of the plurality of classification, from user under the plurality of classification
Middle selection targeted customer;Advertisement putting is carried out to the targeted customer.As can be seen here, with interest-degree, novelty degree and quality three
With reference to mode go calculate classification fraction so that the fraction of the classification for calculating is more accurate, is not to choose whole label successively
Under user, but first using the best result in the fraction of all categories belonging to each user as user fraction, according to each
The fraction of individual user order from big to small is ranked up to each user, throws in crowd further according to advertising presupposition and measures directly from institute
Targeted customer is chosen in each user for stating sequence, the targeted customer of selection is more accurate, and selection process take it is short, so as to reality
Determine that targeted customer carries out advertisement putting with having showed fast accurate.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply these entities or deposit between operating
In any this actual relation or order.Term " including ", "comprising" or its any other variant are intended to non-row
His property is included, so that a series of process, method, article or equipment including key elements not only include those key elements, and
And also include other key elements being not expressly set out, or also include for this process, method, article or equipment institute inherently
Key element.In the absence of more restrictions, the key element for being limited by sentence "including a ...", it is not excluded that including institute
Also there is other identical element in process, method, article or the equipment of stating key element.
For device embodiment, because it corresponds essentially to embodiment of the method, so related part is referring to method reality
Apply the part explanation of example.System embodiment described above is only schematic, wherein described as separating component
The unit of explanation can be or may not be physically separate, can be as the part that unit shows or can also
It is not physical location, you can be located at a place, or can also be distributed on multiple NEs.Can be according to reality
Need the purpose for selecting some or all of module therein to realize this embodiment scheme.Those of ordinary skill in the art are not
In the case of paying creative work, you can to understand and implement.
The above is only the specific embodiment of the application, it is noted that for the ordinary skill people of the art
For member, on the premise of without departing from the application principle, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as the protection domain of the application.
Claims (10)
1. a kind of method of advertisement putting, it is characterised in that include:
The seed user in history of advertising data is obtained, the seed user is the user for completely watching the advertisement;
With interest-degree, novelty degree and quality as parameter, the fraction of multiple classifications of video user, institute are calculated by Similarity Algorithm
The ratio that interest-degree refers to seed user under the classification is stated, the novel degree refers to the ratio of other users under the classification,
The quality refers to the ratio of seed user under classification described in preset number of days historical data, and the other users are sowed for classification
User beyond child user;
According to the fraction of the plurality of classification, targeted customer is chosen in user under the plurality of classification;
Advertisement putting is carried out to the targeted customer.
2. method according to claim 1, it is characterised in that the formula that the Similarity Algorithm is calculated is:Score=a
× ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction, and ln is natural logrithm, institute
It is interest-degree to state interest, and the novelty is novelty degree, and the quality is quality, and a, b, c are predetermined coefficient.
3. method according to claim 1, it is characterised in that the interest-degree is equal to the seed user and the classification
The ratio of the common factor number of users of lower user and number of users under the classification, the novel degree subtracts the interest-degree, institute equal to 1
Quality is stated equal to seed user quantity under classification described in preset number of days historical data and class described in preset number of days historical data
Do not descend the ratio of number of users.
4. method according to claim 1, it is characterised in that the fraction according to the plurality of classification, from described many
Targeted customer is chosen in user under individual classification, including:
Using the best result in the fraction of all categories belonging to each user as user fraction, according to the fraction of each user
Order from big to small is ranked up to each user;
Targeted customer is chosen in each user sorted under the plurality of classification.
5. method according to claim 1, it is characterised in that the fraction according to the plurality of classification, from described many
Targeted customer is chosen in user under individual classification, specially:Crowd's amount is thrown according to the fraction and advertising presupposition of the plurality of classification,
Targeted customer is chosen under the plurality of classification in user.
6. a kind of device of advertisement putting, it is characterised in that include:
Acquiring unit, for obtaining history of advertising data in seed user, the seed user is completely to watch the advertisement
User;
Computing unit, for interest-degree, novelty degree and quality as parameter, by Similarity Algorithm the multiple of video user being calculated
The fraction of classification, the interest-degree refers to the ratio of seed user under the classification, and the novel degree refers under the classification it
The ratio of his user, the quality refers to the ratio of seed user under classification described in preset number of days historical data, it is described other
User is the user under classification beyond seed user;
First chooses unit, for according to the fraction of the plurality of classification, choosing target from user under the plurality of classification and using
Family;
Unit is thrown in, for carrying out advertisement putting to the targeted customer.
7. device according to claim 6, it is characterised in that the formula that the Similarity Algorithm is calculated is:Score=a
× ln (interest)+b × ln (novelty)+c × ln (quality), the Score are fraction, and ln is natural logrithm, institute
It is interest-degree to state interest, and the novelty is novelty degree, and the quality is quality, and a, b, c are predetermined coefficient.
8. device according to claim 6, it is characterised in that the interest-degree is equal to the seed user and the classification
The ratio of the common factor number of users of lower user and number of users under the classification, the novel degree subtracts the interest-degree, institute equal to 1
Quality is stated equal to seed user quantity under classification described in preset number of days historical data and class described in preset number of days historical data
Do not descend the ratio of number of users.
9. device according to claim 6, it is characterised in that the first selection unit includes:
Sequencing unit, for using the best result in the fraction of all categories belonging to each user as user fraction, according to
The fraction of each user order from big to small is ranked up to each user;
Second chooses unit, for choosing targeted customer in each user of sequence under the plurality of classification.
10. device according to claim 6, it is characterised in that described first chooses unit specifically for according to described many
The fraction of individual classification and advertising presupposition throw in crowd's amount, and targeted customer is chosen in user under the plurality of classification.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611248934.2A CN106600342A (en) | 2016-12-29 | 2016-12-29 | Advertisement delivery method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611248934.2A CN106600342A (en) | 2016-12-29 | 2016-12-29 | Advertisement delivery method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106600342A true CN106600342A (en) | 2017-04-26 |
Family
ID=58605070
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611248934.2A Pending CN106600342A (en) | 2016-12-29 | 2016-12-29 | Advertisement delivery method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106600342A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108920479A (en) * | 2018-04-16 | 2018-11-30 | 国家计算机网络与信息安全管理中心 | For two micro- across information source account recommended methods in one end |
CN109191217A (en) * | 2018-11-12 | 2019-01-11 | 北京奇艺世纪科技有限公司 | A kind of video ads impressions prediction technique and device |
CN109428928A (en) * | 2017-08-31 | 2019-03-05 | 腾讯科技(深圳)有限公司 | Selection method, device and the equipment of information push object |
WO2020088050A1 (en) * | 2018-10-31 | 2020-05-07 | 北京字节跳动网络技术有限公司 | Information generation method and device |
CN112348587A (en) * | 2020-11-16 | 2021-02-09 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
CN114596126A (en) * | 2022-04-26 | 2022-06-07 | 土巴兔集团股份有限公司 | Advertisement recommendation method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102141986A (en) * | 2010-01-28 | 2011-08-03 | 北京邮电大学 | Individualized information providing method and system based on user behaviors |
CN102411596A (en) * | 2010-09-21 | 2012-04-11 | 阿里巴巴集团控股有限公司 | Information recommendation method and system |
CN102880969A (en) * | 2011-07-13 | 2013-01-16 | 阿里巴巴集团控股有限公司 | Advertisement putting method, advertisement putting server and advertisement putting system |
CN105550903A (en) * | 2015-12-25 | 2016-05-04 | 腾讯科技(深圳)有限公司 | Target user determination method and apparatus |
-
2016
- 2016-12-29 CN CN201611248934.2A patent/CN106600342A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102141986A (en) * | 2010-01-28 | 2011-08-03 | 北京邮电大学 | Individualized information providing method and system based on user behaviors |
CN102411596A (en) * | 2010-09-21 | 2012-04-11 | 阿里巴巴集团控股有限公司 | Information recommendation method and system |
CN102880969A (en) * | 2011-07-13 | 2013-01-16 | 阿里巴巴集团控股有限公司 | Advertisement putting method, advertisement putting server and advertisement putting system |
CN105550903A (en) * | 2015-12-25 | 2016-05-04 | 腾讯科技(深圳)有限公司 | Target user determination method and apparatus |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109428928A (en) * | 2017-08-31 | 2019-03-05 | 腾讯科技(深圳)有限公司 | Selection method, device and the equipment of information push object |
CN109428928B (en) * | 2017-08-31 | 2021-01-05 | 腾讯科技(深圳)有限公司 | Method, device and equipment for selecting information push object |
CN108920479A (en) * | 2018-04-16 | 2018-11-30 | 国家计算机网络与信息安全管理中心 | For two micro- across information source account recommended methods in one end |
CN108920479B (en) * | 2018-04-16 | 2022-06-17 | 国家计算机网络与信息安全管理中心 | Cross-information-source account recommendation method for two micro terminals |
WO2020088050A1 (en) * | 2018-10-31 | 2020-05-07 | 北京字节跳动网络技术有限公司 | Information generation method and device |
CN109191217A (en) * | 2018-11-12 | 2019-01-11 | 北京奇艺世纪科技有限公司 | A kind of video ads impressions prediction technique and device |
CN112348587A (en) * | 2020-11-16 | 2021-02-09 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
CN112348587B (en) * | 2020-11-16 | 2024-04-23 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
CN114596126A (en) * | 2022-04-26 | 2022-06-07 | 土巴兔集团股份有限公司 | Advertisement recommendation method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106600342A (en) | Advertisement delivery method and device | |
CN108230024B (en) | Advertisement putting engine system based on clue collection | |
US20160299943A1 (en) | Search query refinement | |
US8893012B1 (en) | Visual indicator based on relative rating of content item | |
US8321496B2 (en) | User evaluation of content on distributed communication network | |
JP2006522963A (en) | Method and apparatus for determining a minimum cost per click for a term in an auction based internet search | |
Lommatzsch | Real-time news recommendation using context-aware ensembles | |
CN107368488A (en) | A kind of method for determining user behavior preference, the methods of exhibiting and device of recommendation information | |
US20040153360A1 (en) | System and method of message selection and target audience optimization | |
US20070078758A1 (en) | Risk assessment-based shopping of internet auctions | |
KR20090089919A (en) | Network node ad targeting | |
US20080270364A1 (en) | Expansion rule evaluation | |
CA2710238A1 (en) | Video quality measures | |
DE10244974A1 (en) | Automatic advertiser notification for a system for providing place and price protection in a search result list generated by a computer network search engine | |
CN105654198B (en) | Brand advertisement effect optimization method capable of realizing optimal threshold value selection | |
CN104750783A (en) | Advertisement putting method | |
US20190050890A1 (en) | Video dotting placement analysis system, analysis method and storage medium | |
France et al. | Characterizing viral videos: Methodology and applications | |
US8868565B1 (en) | Calibrating click duration according to context | |
CN109598544A (en) | A kind of media object matching process and system | |
CN110942376A (en) | Fusion method of real-time multi-recall strategy of audio products | |
CN106649323A (en) | Method and device for recommending keyword | |
AU2010253947A1 (en) | An electronic advertisement system | |
CN113407816A (en) | Object recommendation method and device, electronic equipment and storage medium | |
CN113158032B (en) | Information pushing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170426 |