CN103365900A - Method and device for throwing on-line material - Google Patents
Method and device for throwing on-line material Download PDFInfo
- Publication number
- CN103365900A CN103365900A CN2012100966107A CN201210096610A CN103365900A CN 103365900 A CN103365900 A CN 103365900A CN 2012100966107 A CN2012100966107 A CN 2012100966107A CN 201210096610 A CN201210096610 A CN 201210096610A CN 103365900 A CN103365900 A CN 103365900A
- Authority
- CN
- China
- Prior art keywords
- line
- upper channel
- matching degree
- line upper
- described line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention discloses a method and device for throwing on-line material. The method for throwing the on-line material includes the steps that the matching degree of the on-line material and an on-line channel is calculated, and whether the matching degree is larger than a preset threshold value or not is judged; if the matching degree is larger than the preset threshold value, the on-line material is thrown on the on-line channel. Due to the method and device for throwing the on-line material, the on-line material can be thrown on the appropriate on-line channel conveniently.
Description
Technical field
The application relates to data processing field, material put-on method and device on a kind of line.
Background technology
Material is thrown in and is referred to by the material advertisement putting is reached for the purpose of material advertisement putting side introducing flow to the website of outside partner.
Fig. 1 is the synoptic diagram according to the material put-on method of prior art.As shown in Figure 1, the method comprises:
Step S1, business development (Business Development, referred to as BD) personnel are according to ad material on the creative making line, wherein, BD personnel refer to do business expansion, business developing, be responsible for the outer personnel that throw of advertisement on the line specially, advertisement refers to use network to carry out the advertising campaign form as carrier on the line, and material comprises the content, form of advertisement etc.;
Step S2, the BD personnel arrive the cooperation website according to the experience of oneself with advertisement putting;
Step S3, in the process of throwing in, data warehouse (Data Warehouse, referred to as DW) gather user behavior data that partner site's drainage comes, the effect monitoring form is provided, for material is thrown on line upper channel more suitably, the BD personnel can carry out some optimizations and adjustment with reference to these user behavior datas and effect monitoring form.
Can find out from above-mentioned flow process: DW gathers and has stored the user behavior historical data, and BD personnel main reference effect monitoring form comes effect of optimization according to experience to throwing in material and channel adjustment, but user behavior data effectively do not utilized; Along with business development, ad material, cooperation channel get more and more, operation becomes increasingly complex, attends to one thing and lose sight of another easily, how to allow material render to suitable channel to reach optimum efficiency, it is the more and more problem of headache of BD and operation personnel, wherein, channel refers to the line upper channel that is used for bearing advertisement that outside partner provides.
Be difficult to be devoted to the problem of suitable line upper channel for material on the prior art center line, not yet propose at present effective solution.
Summary of the invention
The application's fundamental purpose is to provide material put-on method and device on a kind of line, is difficult to be devoted to the problem of suitable line upper channel with material in the solution prior art.
To achieve these goals, an aspect according to the application provides material put-on method on a kind of line.The material put-on method comprises on this line: the matching degree of material and line upper channel on the calculating line; Judge that whether matching degree is greater than predetermined threshold; And, material on the line is rendered on the line upper channel during greater than predetermined threshold in matching degree.
Further, the matching degree of material and line upper channel comprises on the calculating line: the input service data by the user is carried out simulated training, obtains data model; And the matching degree of calculating material and line upper channel on the line by data model.
Further, the matching degree of material and line upper channel comprises on the calculating line: the line upper channel is classified; Calculate the line upper channel preference level of significance of material on the line by data model; And the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line.
Further, the line upper channel is classified comprise: different line upper channels is stamped different labels so that the line upper channel is classified.
Further, determine that according to the line upper channel preference level of significance of material on the line matching degree of material and line upper channel comprises on the line: data model is fed back to data warehouse, and wherein, data warehouse is used for material on the storage line; Obtain material on data model and the line by carrying out data interaction with data warehouse; And the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line that calculates.
Further, according to matching degree material on the line being rendered to the line upper channel comprises: determine with line on the material matching degree greater than one or more line upper channels of predetermined threshold; And material on the line rendered on definite line upper channel.
Further, with line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, said method comprises: will sort with a plurality of line upper channels of material matching degree on the line greater than predetermined threshold; And with material on the line according to the ordering the result render to successively on a plurality of line upper channels.
To achieve these goals, according to the application on the other hand, provide material delivery device on a kind of line.The material delivery device comprises on this line: computing module, for the matching degree of material and line upper channel on the calculating line; Judge module is used for judging that whether matching degree is greater than predetermined threshold; And putting module,, material on the line is rendered on the line upper channel during greater than predetermined threshold in matching degree.
Further, computing module comprises: the modeling submodule, be used for carrying out simulated training by user's input service data, and obtain data model; And calculating sub module, be used for the matching degree by material and line upper channel on the data model calculating line.
Further, computing module also comprises: the classification submodule is used for the line upper channel is classified; Wherein, calculating sub module is used for calculating by data model the line upper channel preference level of significance of material on the line, and the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line.
Further, the classification submodule is used for by different line upper channels is stamped different labels so that the line upper channel is classified.
Further, computing module also comprises: the feedback submodule, be used for data model is fed back to data warehouse, wherein, data warehouse is used for material on the storage line, wherein, calculating sub module is used for obtaining material on data model and the line by carrying out data interaction with data warehouse, and the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line that calculates.
Further, putting module comprises: determine submodule, be used for to determine with line on the material matching degree greater than one or more line upper channels of predetermined threshold; And the input submodule, be used for material on the line is rendered to definite line upper channel.
Further, putting module also comprises: order module, be used for line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, to sort with a plurality of line upper channels of material matching degree on the line greater than predetermined threshold, wherein, throwing in submodule is used for material on the line is rendered to a plurality of line upper channels successively according to the result who sorts.
By the application, the matching degree of material and line upper channel judges that whether matching degree is greater than predetermined threshold on the employing calculating line; And in matching degree during greater than predetermined threshold, material on the line is rendered to method on the line upper channel, owing to automatically material on the line is rendered on the line upper channel by the matching degree of material on the line that calculates and line upper channel, thereby solved that material is difficult to be devoted to the problem of suitable line upper channel on the prior art center line, and then reached and make material on the line render to easily effect on the suitable line upper channel.
Description of drawings
The accompanying drawing of formation the application's a part is used to provide the further understanding to the application, and the application's illustrative examples and explanation thereof are used for explaining the application, do not consist of the improper restriction to the application.In the accompanying drawings:
Fig. 1 is the synoptic diagram according to material put-on method on the line of prior art;
Fig. 2 is the synoptic diagram according to material delivery device on the line of the application the first embodiment;
Fig. 3 is the synoptic diagram according to material delivery device on the line of the application the second embodiment;
Fig. 4 is the synoptic diagram according to material delivery device on the line of the application the 3rd embodiment;
Fig. 5 is the functional framework figure that throws in according to material on the line of the application's a preferred embodiment;
Fig. 6 is the tree-shaped synoptic diagram in the tag along sort storehouse that provides according to the embodiment of the present application;
Fig. 7 is the synoptic diagram according to material delivery device on the line of the application the 4th embodiment;
Fig. 8 is the synoptic diagram according to material delivery device on the line of the application the 5th embodiment; And
Fig. 9 is the process flow diagram according to material put-on method on the line of the embodiment of the present application.
Embodiment
Need to prove that in the situation of not conflicting, embodiment and the feature among the embodiment among the application can make up mutually.Describe below with reference to the accompanying drawings and in conjunction with the embodiments the application in detail.
Embodiment 1
Fig. 2 is the synoptic diagram according to material delivery device on the line of the application the first embodiment.
As shown in Figure 2, the material delivery device comprises on this line:
Computing module 10 is for the matching degree of material and line upper channel on the calculating line;
Judge module 20 is used for judging that whether described matching degree is greater than predetermined threshold; And
Putting module 30, is rendered to material on the described line on the described line upper channel during greater than described predetermined threshold in described matching degree.
On this line in the material delivery device, because after the matching degree by material and line upper channel on the computing module 10 calculating lines, material and the matching degree of line upper channel and the judged result of judge module 30 are rendered to material on the line on the line upper channel automatically on the line that putting module 30 can calculate by computing module 10, thus can so that on the line material render to easily on the suitable line upper channel.
The application is pointed renders to material on the line on the line upper channel can be that form by Internet Transmission transfers to the line upper channel with material, like this, when the material delivery device is thrown on utilizing the application's line, can only give the line upper channel higher with its matching degree with the material advertisement by network delivery, and material advertisement delivery that can matching degree is not lower is to the line upper channel, thereby reduced the data volume of material advertisement transmission in the network, improve the speed of data transmission, avoided the excessive network congestion that causes of data volume in the network.
Embodiment 2
For so that computing module 10 calculates the matching degree of material on the line and line upper channel more accurately, the embodiment of the present application also provides material delivery device on a kind of preferred line, as shown in Figure 3.
Fig. 3 is the synoptic diagram according to material delivery device on the line of the application the second embodiment.As shown in the figure, in this embodiment, the material delivery device comprises computing module 10 and putting module 30 on this line, wherein, computing module 10 comprises: modeling submodule 101 is used for by user behavior data (perhaps, user's input service data) carries out simulated training, obtain data model; And calculating sub module 102, be used for the matching degree by material and line upper channel on the data model calculating line.Wherein, user behavior data can be that e-commerce website has been thrown in after the advertisement, attract some users to browse e-commerce website, and carry out the movable data that produce at e-commerce website, user behavior data has comprised the c00kie of calling party, the material of access, the passage of access etc., the user is when carrying out material input operation, for example, when throwing in advertising campaign by e-commerce platform to number of site, can rule of thumb material on the line be rendered on the suitable line upper channel, thereby, the modeling submodule 101 of the embodiment of the present application is by user's operation, namely obtain the corresponding relation of material and line upper channel on the line, thereby can obtain the data module about material on the line and line upper channel matching relationship.
In this embodiment, input service data by 101 couples of users of modeling submodule is carried out simulated training, obtain data model, and the matching degree of calculating material and line upper channel on the line by data model, because data model is to carry out simulated training by the data of the user being thrown in operation to obtain, thereby, when calculating matching degree by this model, can more be close to user's practical operation, thereby improve the accuracy that matching degree is calculated.
The material delivery device can be realized at e-commerce platform on the line among the application, throws in ad material (material on the line) to reach the purpose of throwing in advertisement to other websites by e-commerce platform to other website.
Embodiment 3
Fig. 4 is the synoptic diagram according to material delivery device on the line of the application the 3rd embodiment.
For so that calculating sub module 102 is calculated the matching degree of material and line upper channel on the line more easily by data model, preferably, computing module 10 can also comprise: classification submodule 103 is used for the line upper channel is classified; Wherein, calculating sub module 102 is used for calculating by data model the line upper channel preference level of significance of material on the line, and the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line.Wherein, the preference level of significance is used for reflecting that material is to line upper channel matching degree on the line, and the preference level of significance is higher, and the expression matching degree is higher.
In this embodiment, classify by 103 pairs of line upper channels of classification submodule, the line upper channel is divided into different types, thereby, when material mates on dissimilar line upper channels and the line, can carry out according to type the calculating of matching degree, and then so that when the line upper channel preference level of significance of material is determined the matching degree of material and line upper channel on the line on according to line, calculate easier.
The embodiment of the present application also provides a kind of preferred implementation that the line upper channel is classified, that is, make classification submodule 103 come the line upper channel is classified by different line upper channels is stamped different labels.
Fig. 5 is the functional framework figure that throws in according to material on the line of the application's a preferred embodiment.
In this functional framework, by channel is carried out tagging management, and external site introducing user's behavioral data carries out data modeling on this basis, calculate the preference label of outer throwing channel, task engine is rendered to suitable channel according to the preference label of channel with material, accomplish data-driven, before thoroughly changing BD staff by rule of thumb, the working method of people's meat.Wherein, channel namely refers to the line upper channel for material on the lifting-line.
Whole functional framework can be thrown in scene by tag library system, data warehouse, algorithm model, task engine, material and form.In detail as shown in the figure:
1, label is classified, stamped to channel, and label system call number is processed to carry out data according to the warehouse.
As mentioned above, classification submodule 103 can come the line upper channel is classified by different line upper channels is stamped different labels, with label system call number according to the warehouse after, i.e. corresponding relation by line upper channel and label and then obtained the corresponding relation of material and label on the line, wherein, data warehouse is used for material on the storage line.
2, data modeling, algorithm model are trained, and calculate the channel label preference level of significance of material; Model feedback is to data warehouse simultaneously.
Thereby, above-mentioned computing module 10 can also comprise: feedback submodule 103, be used for data model is fed back to data warehouse, at this moment, calculating sub module can be by carrying out the matching degree that data interaction determines according to the line upper channel preference level of significance of material on the line that calculates material and line upper channel on the line with data warehouse.Because material and data model all are stored in the data warehouse on the line, thereby, in the line upper channel preference level of significance of calculating material on the line, equal calling datas from data warehouse, thus improved the speed of calculating.
3, task engine and data warehouse are mutual, carry out the task configuration according to the material channel preference level of significance that data model calculates, that is, be configured according to the put-on method of the matching degree that calculates to material on the line;
4, task engine carries out advertisement and throws outward according to configuring of task;
5, user behavior data collects data warehouse, forms closed loop.
Below label system, data that the embodiment of the present application is provided process, the implementation of data modeling is given an example description.
One, label system
At first, the classification of mentioning in the embodiment of the present application is a kind of relatively-stationary channel attribute description mode, and classification is a kind of tree-shaped institutional framework from data structure, can describe " classification " tissue and refinement from multiple dimension, thereby form many independently classification trees, as shown in Figure 6.
The label that the embodiment of the present application is mentioned is a kind of channel attribute description mode of planar fashion, label take the form of a concrete vocabulary, reflect the attribute of channel by several vocabulary (label).Be fit to throw in " clothing " relevant advertisements material such as channel website A " Chinese costume talent net ", stamp " clothes " label for channel A; Channel website B " China Chemical Industry net " is fit to throw in " chemical " relevant advertisements material, stamps " chemical industry " label for channel B.
Two, data are processed
Computing module 10 by with label system call number according to the warehouse, carry out data and process, data process refer to user behavior data carry out Database Modeling, data gather, and channel identification field on the line are stamped the processing such as label.
2.1, as an example of user behavior data, as shown in the table:
2.2, make up " material and label summary sheet "
Because channel is carried out tagging management, namely, different channels has different labels, and channel refers to the line upper channel for material on the lifting-line, that is, different channels is material on the corresponding different line also, thereby, computing module 10 is by the corresponding relation of channel and label, and the corresponding relation of material and label in the outlet of deriving makes up " material and label summary sheet ":
| Field name | Field description | Data type | Remarks |
| STAT_TIME | Timing statistics | NUMBER | Timing statistics |
| LOCATION_ID | Material ID | NUMBER | BD throws material ID outward |
| TAG_ID | Label ID | NUMBER | Label ID |
| VISIT_NUM | Browse index | NUMBER | As: access times, the number of visiting people |
| SRH_NUM | Search target | NUMBER | As: searching times, search number |
| FB_NUM | The feedback index | NUMBER | As: Times of Feedback, feedback number |
| ORDER_NUM | The transaction index | NUMBER | As: lower single index, payment index |
| ...... | ...... | ...... | ...... |
Three, data modeling
On the basis that the second portion data are processed, by data digging method and business experience user behavior data is set up rule model, rule is configurable, can edit.
3.1, make up average/variance parameter table
Calculate average and two indexs of variance of user behavior statistical indicator for the record of material among each the group TAG_ID in material and the label summary sheet, for example: number of visits, transaction count.Computing method: computing module 10 takes out the record (for example, transaction index, index that browse, feedback be sky such as record value, or data are 0, then ignores this record) of all materials among each TAG; Material on the line is carried out ascending order to be arranged; Ignore maximum 2% and minimum 2% record; Ask average and the variance of this desired value in the last record.Its structure can be with reference to following table:
3.2, Z-SCORE method standardized data
Computing module 10 utilizes the z-score algorithm that the data of material on the line (comprise all user behavior statistical indicators, as: number of visits, transaction count) are carried out standardization.
This method is carried out the standardization of data based on average (mean) and the standard deviation (standard deviation) of the raw data of material on the line.
The computing method of using z-score to be normalized into v ' the original value v of the attribute A of user behavior statistical indicator are:
Wherein,
Be the average of attribute A original value, average can directly read the average of corresponding field.σ
AThe standard deviation of attribute A original value, standard deviation sigma
ABe variances sigma
2Square root, the average that TAG is corresponding and variance data can directly read.
3.3, extreme value processes
Each relative value of 10 pairs of user behavior datas of computing module, value is greater than 2 or replace with 2 ,-2 respectively less than-2 value.
3.4, Min-max method standardized data
The min-max standardized method is the linear transformation to raw data.
If minA and maxA are respectively minimum value and maximal value in the original value of attribute A, the original value v of attribute A is mapped in interval [0,1] by the min-max standardization, then formula of reduction is:
User behavior data is relatively fixing through front conversion value several times, and its interval is [2,2], can directly adopt following formula normalized value:
3.5, calculate material label preference
Computing module 10 quantizes the preference degree (that is, line on the coupling probability of material and certain label) of material on the line for certain label by certain computing method.
Definition: S is level of significance, S
jBe the sign material level of significance (the probability size of material on certain label on the line) on certain label j of material on line, its value is by the corresponding n of a label j factor (material on each line of containing under the label j) F
jWith weight a
jConsist of.Material preference level of significance computing formula:
S
ji＝a
j1*F
j1+a
j2*F
j2+…+a
jn*F
jn
Through the calculating of conspicuousness formula, generate level of significance table R0, as shown in the table:
The computing method of conspicuousness coefficient, set up the reference data sample by the off-line sampling, on the basis of sampling sample, by analytical approachs such as LgisticRegression, each class behavior is carried out match to the weighted value of the impact of level of significance, analysis result (weight coefficient on the level of significance impact of each class behavior) is input in the system, and task engine is finished the quantification of the level of significance of whole materials is calculated.Material label preference historical information is integrated, is generated the wide table of material label preference:
After material is stamped label on to line, label is namely as an attribute of material on the line, thereby can obtain by the preference level of label the line upper channel preference level of significance of material on the line, at this moment, putting module 30, can be chosen the higher several channels of material preference level of significance and throw in when carrying out the material input based on material label preference level, improve input-output ratio (Return On Investment is referred to as ROI).How task engine mates and can be configured with reference to the value of the preference level of significance of material material, channel, realizes that business scenario is thrown in the material robotization of data-driven.
Fig. 7 is the synoptic diagram according to material delivery device on the line of the application the 4th embodiment.
For so that the mechanism of throwing in is more flexible, putting module 30 can comprise:
Determine submodule 201, be used for to determine with line on the material matching degree greater than one or more line upper channels of predetermined threshold; And
Throw in submodule 202, be used for material on the line is rendered to definite line upper channel.
In this embodiment, determine that submodule 201 determines with line on the material matching degree be one or more greater than the line upper channel of predetermined threshold, with line on material matching degree when being greater than the line upper channel of predetermined threshold, can be to material on the automatic switching unwrapping wire on the line upper channel of determining, with line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, can be on the determinate multitudes line upper channel material on the automatic switching unwrapping wire simultaneously.
Fig. 8 is the synoptic diagram according to material delivery device on the line of the application the 5th embodiment.
For so that the mechanism of throwing in is more flexible, putting module also comprises:
Order module 203, be used for line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, will sort with a plurality of line upper channels of material matching degree on the line greater than predetermined threshold,
Wherein, throwing in submodule is used for material on the line is rendered to a plurality of line upper channels successively according to the result who sorts.
In this embodiment, realized so that throw in submodule can according to the ordering the result successively material on the line is rendered on a plurality of line upper channels.
The embodiment of the present application also provides material put-on method on a kind of line, and the material put-on method can adopt on the line that provides in above-described embodiment material to throw in to realize on this line.
Fig. 9 is the process flow diagram according to material put-on method on the line of the embodiment of the present application.As shown in the figure, the method comprises:
Step S602, the matching degree of material and line upper channel on the calculating line;
Step S604 judges that whether matching degree is greater than predetermined threshold; And
Step S606, renders to material on the line on the line upper channel during greater than predetermined threshold in matching degree.
By material put-on method on this line, because after the matching degree by material and line upper channel on the calculating line, can automatically material on the line be rendered on the line upper channel by the matching degree of material on the line that calculates and line upper channel, thus can so that on the line material render to easily on the suitable line upper channel.
As a kind of preferred implementation of calculating the matching degree of material and line upper channel on the line, can adopt following methods to calculate the matching degree of material and line upper channel on the line: the input service data by the user is carried out simulated training, obtains data model; And the matching degree of calculating material and line upper channel on the line by data model.
In this embodiment, carry out simulated training by the input service data to the user, obtain data model, and the matching degree of calculating material and line upper channel on the line by data model, because data model is to carry out simulated training by the data of the user being thrown in operation to obtain, thereby, when calculating matching degree by this model, can more be close to user's practical operation, thereby improve the accuracy that matching degree is calculated.
Preferably, can adopt following methods to calculate the matching degree of material and line upper channel on the line: the line upper channel is classified; Calculate the line upper channel preference level of significance of material on the line by data model; And the matching degree of determining material and line upper channel on the line according to the line upper channel preference level of significance of material on the line.
Matching degree by material and line upper channel on employing the method calculating line, because the line upper channel is classified, the line upper channel is divided into different types, thereby, when material mates on dissimilar line upper channels and the line, can carry out according to type the calculating of matching degree, and then so that when the line upper channel preference level of significance of material is determined the matching degree of material and line upper channel on the line on according to line, calculate easier.
As a kind of preferred implementation that the line upper channel is classified, can be by different line upper channels being stamped different labels so that the line upper channel is classified to realize the line upper channel is carried out Fast Classification.
Corresponding to aforesaid device embodiment, (material is higher in certain line upper channel preference level of significance on the line to determine the matching degree of material and line upper channel on the line according to the line upper channel preference level of significance of material on the line, illustrate that then the probability of material on this line upper channel is higher on the line, the matching degree of the two is also higher, vice versa) comprising: data model is fed back to data warehouse, wherein, data warehouse is used for material on the storage line; And by carrying out the matching degree that data interaction determines according to the line upper channel preference level of significance of material on the line that calculates material and line upper channel on the line with data warehouse.At this moment, because material all is stored in the data warehouse on data model and the line, thereby more conveniently call material on the line by data model.
Preferably, according to matching degree material on the line being rendered to the line upper channel comprises: determine with line on the material matching degree greater than one or more line upper channels of predetermined threshold; And material on the line rendered on definite line upper channel, thereby so that on line material can render to and the one or more line upper channels of its matching degree greater than predetermined threshold, realized the flexible input of material on the line.
With line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, said method can also comprise: will sort with a plurality of line upper channels of material matching degree on the line greater than predetermined threshold; And with material on the line according to the ordering the result render to successively on a plurality of line upper channels, thereby realized the orderly input of material on the line.
Obviously, those skilled in the art should be understood that, each module of above-mentioned the application or each step can realize with general calculation element, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation elements form, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in the memory storage and be carried out by calculation element, perhaps they are made into respectively each integrated circuit modules, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the application is not restricted to any specific hardware and software combination.
The above is the application's preferred embodiment only, is not limited to the application, and for a person skilled in the art, the application can have various modifications and variations.All within the application's spirit and principle, any modification of doing, be equal to replacement, improvement etc., all should be included within the application's the protection domain.
Claims (14)
1. material put-on method on the line is characterized in that comprising:
Calculate the matching degree of material and line upper channel on the described line;
Judge that whether described matching degree is greater than predetermined threshold; And
, material on the described line is rendered on the described line upper channel during greater than described predetermined threshold in described matching degree.
2. material put-on method according to claim 1 is characterized in that, the matching degree of material and line upper channel comprises on the calculating line:
Input service data by the user is carried out simulated training, obtains data model; And
Calculate the matching degree of material and line upper channel on the described line by described data model.
3. material put-on method according to claim 2 is characterized in that, calculates that the matching degree of material and line upper channel comprises on the described line:
Described line upper channel is classified;
Calculate the line upper channel preference level of significance of material on the described line by described data model; And
Determine the matching degree of material and described line upper channel on the described line according to the line upper channel preference level of significance of material on the described line.
4. material put-on method according to claim 3 is characterized in that, described line upper channel is classified to be comprised:
Different line upper channels is stamped different labels so that described line upper channel is classified.
5. material put-on method according to claim 3 is characterized in that, determines that according to the line upper channel preference level of significance of material on the described line matching degree of material and described line upper channel comprises on the described line:
Described data model is fed back to data warehouse, and wherein, described data warehouse is used for storing material on the described line;
Obtain material on described data model and the described line by carrying out data interaction with described data warehouse; And
Determine the matching degree of material and described line upper channel on the described line according to the line upper channel preference level of significance of material on the described line that calculates.
6. material put-on method according to claim 1 is characterized in that, according to described matching degree material on the described line is rendered to described line upper channel and comprises:
Determine with described line on the material matching degree greater than the one or more described line upper channel of described predetermined threshold; And
Material on the described line is rendered on definite line upper channel.
7. material put-on method according to claim 6 is characterized in that, with described line on material matching degree when being a plurality of greater than the line upper channel of predetermined threshold, described method comprises:
To sort with a plurality of described line upper channel of material matching degree on the described line greater than described predetermined threshold; And
The result of material on the described line according to ordering rendered on a plurality of described line upper channels successively.
8. material delivery device on the line is characterized in that comprising:
Computing module is used for calculating the matching degree of material and line upper channel on the described line;
Judge module is used for judging that whether described matching degree is greater than predetermined threshold; And
Putting module, is rendered to material on the described line on the described line upper channel during greater than described predetermined threshold in described matching degree.
9. material delivery device on the line according to claim 8 is characterized in that, described computing module comprises:
The modeling submodule is used for carrying out simulated training by user's input service data, obtains data model; And
Calculating sub module is used for calculating by described data model the matching degree of material and line upper channel on the described line.
10. material delivery device on the line according to claim 9 is characterized in that, described computing module also comprises:
The classification submodule is used for described line upper channel is classified;
Wherein, described calculating sub module is used for calculating by described data model the line upper channel preference level of significance of material on the described line, and the matching degree of determining material and described line upper channel on the described line according to the line upper channel preference level of significance of material on the described line.
11. the material delivery device is characterized in that on the line according to claim 10, described classification submodule is used for by different line upper channels being stamped different labels so that described line upper channel is classified.
12. the material delivery device is characterized in that on the line according to claim 10, described computing module also comprises:
The feedback submodule is used for described data model is fed back to data warehouse, and wherein, described data warehouse is used for storing material on the described line,
Wherein, described calculating sub module is used for obtaining material on described data model and the described line by carrying out data interaction with described data warehouse, and the matching degree of determining material and described line upper channel on the described line according to the line upper channel preference level of significance of material on the described line that calculates.
13. the material delivery device is characterized in that on the line according to claim 8, described putting module comprises:
Determine submodule, be used for to determine with described line on the material matching degree greater than the one or more described line upper channel of predetermined threshold; And
Throw in submodule, be used for material on the described line is rendered to definite line upper channel.
14. the material delivery device is characterized in that on the line according to claim 13, described putting module also comprises:
Order module, be used for described line on material matching degree when being a plurality of greater than the line upper channel of described predetermined threshold, will sort with a plurality of described line upper channel of material matching degree on the described line greater than predetermined threshold,
Wherein, described input submodule is used for material on the described line is rendered to a plurality of described line upper channels successively according to the result who sorts.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210096610.7A CN103365900B (en) | 2012-04-01 | 2012-04-01 | Material put-on method and device on line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210096610.7A CN103365900B (en) | 2012-04-01 | 2012-04-01 | Material put-on method and device on line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103365900A true CN103365900A (en) | 2013-10-23 |
| CN103365900B CN103365900B (en) | 2018-03-23 |
Family
ID=49367264
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210096610.7A Active CN103365900B (en) | 2012-04-01 | 2012-04-01 | Material put-on method and device on line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103365900B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104299151A (en) * | 2014-09-28 | 2015-01-21 | 新浪网技术(中国)有限公司 | Online advertising flow distribution system and method |
| CN108804430A (en) * | 2017-04-26 | 2018-11-13 | 广东原昇信息科技有限公司 | A kind of SEM launches data sorting system and its sorting technique |
| CN110851653A (en) * | 2019-11-08 | 2020-02-28 | 上海摩象网络科技有限公司 | Method and device for shooting material mark and electronic equipment |
| CN111242654A (en) * | 2018-11-28 | 2020-06-05 | 北京沃东天骏信息技术有限公司 | Method and system for generating advertisement picture |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060218035A1 (en) * | 2003-04-22 | 2006-09-28 | Park Sang W | Method of introducing advertisements and providing the advertisements by using access intentions of internet users and a system thereof |
| CN101290631A (en) * | 2008-05-28 | 2008-10-22 | 北京百问百答网络技术有限公司 | Network advertisement automatic delivery method and its system |
| CN101599160A (en) * | 2009-07-14 | 2009-12-09 | 清华大学 | The method of advertisement delivery and device |
| CN102156721A (en) * | 2011-03-29 | 2011-08-17 | 张栋 | Method for accurately delivering Internet video advertisement based on label |
-
2012
- 2012-04-01 CN CN201210096610.7A patent/CN103365900B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060218035A1 (en) * | 2003-04-22 | 2006-09-28 | Park Sang W | Method of introducing advertisements and providing the advertisements by using access intentions of internet users and a system thereof |
| CN101290631A (en) * | 2008-05-28 | 2008-10-22 | 北京百问百答网络技术有限公司 | Network advertisement automatic delivery method and its system |
| CN101599160A (en) * | 2009-07-14 | 2009-12-09 | 清华大学 | The method of advertisement delivery and device |
| CN102156721A (en) * | 2011-03-29 | 2011-08-17 | 张栋 | Method for accurately delivering Internet video advertisement based on label |
Non-Patent Citations (2)
| Title |
|---|
| 张召等: "基于活跃用户特征的论坛广告投放", 《计算机研究与发展》 * |
| 葛汝刚等: "电子商务网站广告投放切换模型及其鲁棒控制", 《东北大学学报（自然科学版）》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104299151A (en) * | 2014-09-28 | 2015-01-21 | 新浪网技术(中国)有限公司 | Online advertising flow distribution system and method |
| CN104299151B (en) * | 2014-09-28 | 2017-11-17 | 新浪网技术(中国)有限公司 | Web advertisement flow distributing system and method |
| CN108804430A (en) * | 2017-04-26 | 2018-11-13 | 广东原昇信息科技有限公司 | A kind of SEM launches data sorting system and its sorting technique |
| CN111242654A (en) * | 2018-11-28 | 2020-06-05 | 北京沃东天骏信息技术有限公司 | Method and system for generating advertisement picture |
| CN111242654B (en) * | 2018-11-28 | 2021-04-30 | 北京沃东天骏信息技术有限公司 | Method and system for generating advertisement picture |
| CN110851653A (en) * | 2019-11-08 | 2020-02-28 | 上海摩象网络科技有限公司 | Method and device for shooting material mark and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103365900B (en) | 2018-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109934619A (en) | User's portrait tag modeling method, apparatus, electronic equipment and readable storage medium storing program for executing | |
| CN103714139B (en) | Parallel data mining method for identifying a mass of mobile client bases | |
| CN107818133B (en) | Residential area network capacity analysis method and system based on big data | |
| CN107451861B (en) | Method for identifying user internet access characteristics under big data | |
| CN103578010A (en) | Method and device generating flow quality comparison parameters and advertisement billing method | |
| CN103345512A (en) | Online advertising click-through rate forecasting method and device based on user attribute | |
| CN106101222A (en) | The method for pushing of information and device | |
| CN106203773A (en) | A kind of method and device of automatic management employee | |
| CN107146089A (en) | The single recognition methods of one kind brush and device, electronic equipment | |
| CN106408329A (en) | Advertisement visitor retrieving method and advertisement putting system | |
| CN105224681B (en) | Customer requirement retrieval method and system based on family's place of working context environmental | |
| CN110033331A (en) | A kind of distribution method of discount coupon, system and terminal device | |
| CN102081774A (en) | Card-raising identification method and system | |
| CN110046174B (en) | population migration analysis method and system based on big data | |
| CN110288350A (en) | User's Value Prediction Methods, device, equipment and storage medium | |
| CN110134845A (en) | Project public sentiment monitoring method, device, computer equipment and storage medium | |
| CN108572988A (en) | A kind of house property assessment data creation method and device | |
| CN105184326A (en) | Active learning multi-label social network data analysis method based on graph data | |
| CN103365900A (en) | Method and device for throwing on-line material | |
| CN104778608A (en) | N+ advertisement putting and optimizing method | |
| CN113706291A (en) | Fraud risk prediction method, device, equipment and storage medium | |
| CN107368501A (en) | The processing method and processing device of data | |
| CN104077288B (en) | Web page contents recommend method and web page contents recommendation apparatus | |
| CN109544271A (en) | A kind of trade managing system | |
| CN109670873A (en) | Real estate opens up objective method, apparatus and server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1186556 Country of ref document: HK |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |