CN109840784A - A kind of ad traffic prediction technique and device - Google Patents
A kind of ad traffic prediction technique and device Download PDFInfo
- Publication number
- CN109840784A CN109840784A CN201711200255.2A CN201711200255A CN109840784A CN 109840784 A CN109840784 A CN 109840784A CN 201711200255 A CN201711200255 A CN 201711200255A CN 109840784 A CN109840784 A CN 109840784A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- keyword
- measured
- traffic
- index table
- 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
Abstract
The invention discloses a kind of ad traffic prediction technique and devices, this method comprises: obtaining the keyword to be measured of advertisement to be measured；Keyword to be measured is compared with the keyword in inverted index table, obtains the reference keyword with Keywords matching to be measured, wherein inverted index table is to establish in advance according to the corresponding historical traffic data of advertisement launched in history；Determine that advertisement corresponding with reference keyword is to refer to ad traffic with reference to advertisement, and by acquisition in inverted index table is corresponding with reference advertisement；It is calculated using reference ad traffic, obtains prediction ad traffic corresponding with advertisement to be measured.Thus, it only needs quickly and easily determine using inverted index table corresponding with advertisement to be measured with reference to advertisement, and then prediction ad traffic is obtained using the reference ad traffic of reference advertisement, to substantially reduce predicted time, the real-time of ad traffic prediction is enhanced, the dispensing strategy for instructing advertisement to be measured is conducive to.
Description
Technical field
The present invention relates to technical field of advertisement is calculated, more specifically to a kind of ad traffic prediction technique and device.
Background technique
Ad traffic refers to the displaying amount of the advertisement in certain time in the case where meeting certain constraint conditions.And ad traffic prediction refers to
In the case where giving certain constraint conditions, the advertising display amount for meeting these constraint conditions within some period in the future is estimated.Advertisement
Volume forecasting is most important for the online allocation strategy in displaying amount contract advertisement, according to the size of ad traffic to advertisement into
Capable online distribution directly affects the effect of advertisement.And in advertising network, it is also desirable to wide according to constraint condition and bid estimation
Displaying amount is accused, with the main carry out decision of auxiliary ad-vertisement.
In the prior art, it is usually launched in data by the history of advertisement and the ad traffic for the advertisement that will be launched is carried out
Prediction, still, the history dispensing data of advertisement are usually very huge, therefore, launch the prediction that data are realized by the history of advertisement
It needs the time long, is unable to satisfy the requirement of advertisement Real time request, i.e., in the prior art to the prediction real-time of ad traffic
It is poor.
In conclusion the poor problem of prediction real-time to ad traffic exists in the prior art.
Summary of the invention
The object of the present invention is to provide a kind of ad traffic prediction technique and devices, existing in the prior art right to solve
The poor problem of the prediction real-time of ad traffic.
To achieve the goals above, the invention provides the following technical scheme:
A kind of ad traffic prediction technique, comprising:
Obtain the keyword to be measured of advertisement to be measured；
The keyword to be measured is compared with the keyword in inverted index table, is obtained and the keyword to be measured
The reference keyword matched, wherein the inverted index table is in advance according to the corresponding historical traffic number of advertisement launched in history
According to foundation；
It is determining with it is described with reference to the corresponding advertisement of keyword be with reference to advertisement, and by acquisition and institute in the inverted index table
It states corresponding with reference to ad traffic with reference to advertisement；
It is calculated using described with reference to ad traffic, obtains prediction ad traffic corresponding with the advertisement to be measured.
Preferably, the keyword to be measured of the advertisement to be measured is obtained, comprising:
Extract keyword corresponding with the advertisement to be measured；
It is matched with the stop words in the deactivated vocabulary that pre-establishes in determining keyword corresponding with the advertisement to be measured
Keyword is failure keyword；
Failure keyword in keyword corresponding with the advertisement to be measured is deleted, keyword to be measured is obtained.
Preferably, the keyword to be measured is compared with the keyword in inverted index table, obtain with it is described to be measured
The reference keyword of Keywords matching, comprising:
Using binary chop, the keyword to be measured is compared with the keyword in the inverted index table, is obtained
To the reference keyword with the Keywords matching to be measured.
Preferably, it is calculated using described with reference to ad traffic, obtains prediction advertisement corresponding with the advertisement to be measured
Flow, comprising:
If the quantity with reference to advertisement is one, it is determined that this is described with reference to the corresponding reference ad traffic of advertisement
The prediction ad traffic of advertisement to be measured；If the quantity with reference to advertisement is multiple, it is determined that each described with reference to advertisement pair
The average value for the reference ad traffic answered is the prediction ad traffic of the advertisement to be measured.
Preferably, the process of inverted index table is established according to the corresponding historical traffic data of advertisement launched in history in advance
Include:
Obtain historical traffic data, the historical traffic data include the advertisement launched in history, launch in history it is every
The corresponding keyword of a advertisement and ad traffic；
Institute is established according to the corresponding relationship between advertisement, keyword and the ad traffic for including in the historical traffic data
State inverted index table.
A kind of ad traffic prediction meanss, comprising:
Module is obtained, for obtaining the keyword to be measured of advertisement to be measured；
Matching module obtains and institute for the keyword to be measured to be compared with the keyword in inverted index table
State the reference keyword of Keywords matching to be measured, wherein the inverted index table is in advance according to the advertisement pair launched in history
What the historical traffic data answered was established；
Determining module, being used for determining advertisement corresponding with the reference keyword is to arrange rope with reference to advertisement, and by described
It is corresponding with the reference advertisement with reference to ad traffic to draw acquisition in table；
Computing module is obtained corresponding pre- with the advertisement to be measured for being calculated using described with reference to ad traffic
Survey ad traffic.
Preferably, obtaining module includes:
Extraction unit, for extracting keyword corresponding with the advertisement to be measured；
Disabling unit, in determining keyword corresponding with the advertisement to be measured and in the deactivated vocabulary that pre-establishes
The matched keyword of stop words is failure keyword；
Unit is deleted, for deleting the failure keyword in keyword corresponding with the advertisement to be measured, is obtained to be measured
Keyword.
Preferably, matching module includes:
Matching unit, for utilizing binary chop, by the key in the keyword to be measured and the inverted index table
Word is compared, and obtains the reference keyword with the Keywords matching to be measured.
Preferably, computing module includes:
Computing unit, if being one for the quantity with reference to advertisement, it is determined that this refers to the corresponding reference of advertisement
Ad traffic is the prediction ad traffic of the advertisement to be measured；If the quantity with reference to advertisement is multiple, it is determined that each
It is described with reference to the corresponding average value with reference to ad traffic of advertisement be the advertisement to be measured prediction ad traffic.
Preferably, further includes:
Module is established, for obtaining historical traffic data, the historical traffic data includes the advertisement launched in history, goes through
The corresponding keyword of each advertisement and ad traffic launched in history；And according to the advertisement for including in the historical traffic data,
Corresponding relationship between keyword and ad traffic establishes the inverted index table.
A kind of ad traffic prediction technique provided by the invention and device, this method comprises: obtaining the to be measured of advertisement to be measured
Keyword；The keyword to be measured is compared with the keyword in inverted index table, is obtained and the keyword to be measured
The reference keyword matched, wherein the inverted index table is in advance according to the corresponding historical traffic number of advertisement launched in history
According to foundation；It is determining with it is described with reference to the corresponding advertisement of keyword be with reference to advertisement, and by being obtained in the inverted index table and
It is described corresponding with reference to ad traffic with reference to advertisement；Calculated using described with reference to ad traffic, obtain with it is described to be measured wide
Accuse corresponding prediction ad traffic.By above-mentioned technical characteristic disclosed in the present application, using advertisement to be measured keyword to be measured by
Determined in the inverted index table established according to the corresponding historical traffic data of the advertisement launched in history it is corresponding with reference to advertisement,
And then by the prediction ad traffic that advertisement to be measured is calculated with reference to ad traffic corresponding with reference advertisement, thus, it is only necessary to
Can quickly and easily be determined using inverted index table it is corresponding with advertisement to be measured with reference to advertisement, and then using with reference to advertisement
Prediction ad traffic is obtained with reference to ad traffic, it is to be measured to estimate without the huge data by obtaining dispensing advertisement in history
The prediction ad traffic of advertisement enhances the real-time of ad traffic prediction, is conducive to refer to substantially reduce predicted time
Lead the dispensing strategy of advertisement to be measured.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow chart of ad traffic prediction technique provided in an embodiment of the present invention；
Fig. 2 is the example diagram that node is supplied in a kind of ad traffic prediction technique provided in an embodiment of the present invention；
Fig. 3 is the example diagram of keyword index in a kind of ad traffic prediction technique provided in an embodiment of the present invention；
Fig. 4 is the example diagram of indexed results in a kind of ad traffic prediction technique provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of ad traffic prediction meanss provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It, can be with referring to Fig. 1, it illustrates a kind of flow chart of ad traffic prediction technique provided in an embodiment of the present invention
The following steps are included:
S11: the keyword to be measured of advertisement to be measured is obtained.
It should be noted that the keyword to be measured of advertisement to be measured can be by the intention of advertisement to be measured, crowd's attribute constraint,
It is extracted in the information such as Regional Property constraint and industry attribute constraint, such as population ages, affiliated region, it specifically can be according to reality
It needs to be determined.
S12: keyword to be measured is compared with the keyword in inverted index table, is obtained and Keywords matching to be measured
With reference to keyword, wherein inverted index table is to establish in advance according to the corresponding historical traffic data of advertisement launched in history.
It should be noted that the keyword for including in inverted index table is key corresponding with the advertisement launched in history
Word, the advertisement that advertisement is as launched in history, the corresponding ad traffic of the advertisement that ad traffic is as launched in history.
S13: determine with reference to the corresponding advertisement of keyword be with reference to advertisement, and by obtained in inverted index table with reference to extensively
It accuses corresponding with reference to ad traffic.
Specifically, it if keyword to be measured is one, can directly be retrieved in inverted index table to be measured with this
The identical keyword of keyword is with reference to keyword, and advertisement corresponding with reference keyword is with reference to advertisement；If key to be measured
Word be it is multiple, then traverse keyword to be measured, distinguish identical key with multiple keywords to be measured by retrieving in inverted index table
Word be with reference to keyword, and include simultaneously all referring to the corresponding advertisement of keyword be with reference to advertisement.
S14: being calculated using reference ad traffic, obtains prediction ad traffic corresponding with advertisement to be measured.
Above-mentioned ad traffic prediction technique disclosed by the invention, using the keyword to be measured of advertisement to be measured by according in history
Determined in the inverted index table that the corresponding historical traffic data of the advertisement of dispensing is established it is corresponding with reference to advertisement, and then by with
With reference to the corresponding prediction ad traffic that advertisement to be measured is calculated with reference to ad traffic of advertisement, thus, it is only necessary to utilize the row's of falling rope
Drawing table can quickly and easily determine corresponding with advertisement to be measured with reference to advertisement, and then utilize the reference ad stream for referring to advertisement
Prediction ad traffic is measured, without launching the huge data of advertisement in history by obtaining to estimate the prediction of advertisement to be measured
Ad traffic enhances the real-time of ad traffic prediction, is conducive to instruct advertisement to be measured to substantially reduce predicted time
Dispensing strategy.
In a kind of ad traffic prediction technique provided in an embodiment of the present invention, the keyword to be measured of advertisement to be measured is obtained, it can
To include:
Extract keyword corresponding with advertisement to be measured；
Determine the matched key of stop words in keyword corresponding with advertisement to be measured and in the deactivated vocabulary pre-established
Word is failure keyword；
Failure keyword in keyword corresponding with advertisement to be measured is deleted, keyword to be measured is obtained.
It should be noted that keyword to be measured here is effective keyword, and effectively keyword refers to not include punctuate
The keyword of the words such as symbol, modal particle, auxiliary word, and deactivating the stop words for including in vocabulary to be includes punctuation mark, the tone
The invalid keyword of the words such as word and auxiliary word can be determined according to actual needs by staff.Stop words is utilized as a result,
Table is filtered the corresponding keyword of advertisement to be measured, and invalid keyword is deleted, and then obtains keyword to be measured, to guarantee
The validity of keyword to be measured further ensures the rapid evaluation for predicting ad traffic.
In addition, keyword to be measured is compared with the keyword in inverted index table, obtain and Keywords matching to be measured
Reference keyword, can specifically include:
Using binary chop, keyword to be measured is compared with the keyword in the inverted index table pre-established,
Obtain the reference keyword with Keywords matching to be measured.
Wherein, binary chop is also known as binary search, specifically: first, it is assumed that element is (i.e. crucial in inverted index table
Word) it is to be arranged by ascending order, the keyword that inverted index table middle position is recorded is compared with keyword to be measured, if the two phase
Deng then searching success；Otherwise it is recorded using middle position and inverted index table is divided into former and later two sublists, if middle position
The keyword of record is greater than keyword to be measured, then further searches for previous sublist, otherwise further search for latter sublist.Repeat with
Upper process, the keyword until finding the condition of satisfaction, makes to search successfully, or until sublist is not present, can not search at this time
Function.Thereby, it is possible to reach faster search speed while number of comparisons is less, further ensures the embodiment of the present invention and mention
A kind of real-time of the ad traffic prediction technique supplied.
In a kind of ad traffic prediction technique provided in an embodiment of the present invention, prediction is calculated using reference ad traffic
The method of ad traffic can be determined according to actual needs, for example each to assign weighted value with reference to ad traffic, using every
It is a with reference to ad traffic and its weighted value to carry out that prediction ad traffic is calculated, it for another example takes with reference to the maximum value in ad traffic
And the average value of minimum value is as prediction ad traffic etc., it is within the scope of the present invention.Preferably, using with reference to wide
It accuses flow to be calculated, obtains prediction ad traffic corresponding with advertisement to be measured, can specifically include:
If the quantity with reference to advertisement is one, it is determined that this is advertisement to be measured with reference to the corresponding reference ad traffic of advertisement
Prediction ad traffic；If the quantity with reference to advertisement is multiple, it is determined that each corresponding with reference to ad traffic with reference to advertisement
Average value be advertisement to be measured prediction ad traffic.
Prediction ad traffic can not only quickly and easily be calculated as a result, additionally it is possible to guarantee prediction ad traffic
Accuracy ensure that a kind of accuracy of ad traffic prediction technique provided in an embodiment of the present invention.
It is corresponding according to the advertisement launched in history in advance in a kind of ad traffic prediction technique provided in an embodiment of the present invention
Historical traffic data establish the process of inverted index table and may include:
Obtain historical traffic data, historical traffic data include the advertisement launched in history, each of launch in history it is wide
Accuse corresponding keyword and ad traffic；
The row's of falling rope is established according to the corresponding relationship between advertisement, keyword and the ad traffic for including in historical traffic data
Draw table.
Specifically, can be by obtaining historical traffic data in launching the dispensing log of advertisement in history, and it will be same wide
A supply node is merged into the displaying accused within the same period, and counting the displaying number of the advertisement during this period of time is to be somebody's turn to do
The ad traffic of node is supplied, as a result, using each supply node as a document of ad traffic prediction reverse indexing, in turn
Inverted index is established to the corresponding document of each supply node, obtains inverted index table.The wherein keyword in inverted index table
The keyword in document is as corresponded to, keyword as corresponding with advertisement can specifically include intention from advertisement, crowd
Attribute constraint, Regional Property constraint, effective keyword in industry attribute constraint, wherein effective keyword can be for using stopping
The keyword extracted is filtered with vocabulary, specifically refers to saying about deactivated vocabulary in above-described embodiment
It is bright.As shown in Fig. 2, for in inverted index table one supply node specific example, inverted index table i.e. be by it is multiple not
What same supply node was formed in form shown in Fig. 2, wherein displaying amount is ad traffic.It is corresponding, as shown in figure 3, being
Inverted index, the i.e. specific example of keyword index are carried out using keyword to be measured, wherein male, Shanghai are key to be measured
Word, positive row's table is inverted index table；And Fig. 4 is the concrete example that the result that inverted index obtains is carried out using keyword to be measured
Son, wherein advertisement is advertisement to be measured, and displaying amount is the ad traffic of corresponding node.It is possible thereby to pass through inverted index table
Quick-searching refers to keyword to corresponding with keyword to be measured, and then is determined by corresponding with reference keyword with reference to advertisement
Its is corresponding with reference to ad traffic out, finally calculates prediction ad traffic by reference ad traffic, above-mentioned with reference to keyword, ginseng
Examining advertisement and referring to keyword is the element in inverted index table.
In addition, in a kind of ad traffic prediction technique provided by the invention, in the ad traffic to be measured for obtaining advertisement to be measured,
And after launching advertisement to be measured, can also include:
It determines the actual advertisement flow of advertisement to be measured, and inverted index table is updated using actual advertisement flow.
Specifically, advertisement to be measured is its practical ad traffic in the displaying number launched in preset time period, and pre-
It, usually can be corresponding with the ad traffic for having included in inverted index table if the period can be determined according to actual needs
Period is identical.Inverted index table is added in advertisement to be measured, keyword to be measured and the actual advertisement flow node new as one
In, to guarantee the real-time and validity of inverted index table.
Corresponding with above method embodiment, the embodiment of the invention also provides a kind of ad traffic prediction meanss, such as Fig. 5
It is shown, may include:
Module 11 is obtained, for obtaining the keyword to be measured of advertisement to be measured；
Matching module 12, for keyword to be measured to be compared with the keyword in inverted index table, obtain with it is to be measured
The reference keyword of Keywords matching, wherein inverted index table is in advance according to the corresponding history stream of advertisement launched in history
Measure what data were established；
Determining module 13, for determining and be with reference to advertisement, and by inverted index table with reference to the corresponding advertisement of keyword
It obtains corresponding with reference to ad traffic with reference to advertisement；
Computing module 14 obtains prediction advertisement corresponding with advertisement to be measured for being calculated using reference ad traffic
Flow.
Above-mentioned ad traffic prediction meanss disclosed by the invention, using the keyword to be measured of advertisement to be measured by according in history
Determined in the inverted index table that the corresponding historical traffic data of the advertisement of dispensing is established it is corresponding with reference to advertisement, and then by with
With reference to the corresponding prediction ad traffic that advertisement to be measured is calculated with reference to ad traffic of advertisement, thus, it is only necessary to utilize the row's of falling rope
Drawing table can quickly and easily determine corresponding with advertisement to be measured with reference to advertisement, and then utilize the reference ad stream for referring to advertisement
Prediction ad traffic is measured, without launching the huge data of advertisement in history by obtaining to estimate the prediction of advertisement to be measured
Ad traffic enhances the real-time of ad traffic prediction, is conducive to instruct advertisement to be measured to substantially reduce predicted time
Dispensing strategy.
In a kind of ad traffic prediction meanss provided in an embodiment of the present invention, obtaining module may include:
Extraction unit, for extracting keyword corresponding with advertisement to be measured；
Disabling unit, for determining deactivating in keyword corresponding with advertisement to be measured and in the deactivated vocabulary pre-established
The matched keyword of word is failure keyword；
Unit is deleted, for deleting the failure keyword in keyword corresponding with advertisement to be measured, obtains key to be measured
Word.
In a kind of ad traffic prediction meanss provided in an embodiment of the present invention, matching module may include:
Matching unit compares keyword to be measured and the keyword in inverted index table for utilizing binary chop
It is right, obtain the reference keyword with Keywords matching to be measured.
In a kind of ad traffic prediction meanss provided in an embodiment of the present invention, computing module may include:
Computing unit, if being one for the quantity with reference to advertisement, it is determined that this is corresponding with reference to advertisement with reference to advertisement
Flow is the prediction ad traffic of advertisement to be measured；If the quantity with reference to advertisement is multiple, it is determined that each corresponding with reference to advertisement
Reference ad traffic average value be advertisement to be measured prediction ad traffic.
In a kind of ad traffic prediction meanss provided in an embodiment of the present invention, can also include:
Module is established, for obtaining historical traffic data, historical traffic data includes the advertisement launched in history, in history
The corresponding keyword of each advertisement and ad traffic launched；And according to include in historical traffic data advertisement, keyword and
Corresponding relationship between ad traffic establishes inverted index table.
It should be noted that can use the unit for being filtered to invalid keyword, i.e. disabling unit and deletion
Unit is filtered the failure keyword for including in historical traffic data, can also be by establishing module for historical traffic data
In include failure keyword be filtered, it is within the scope of the present invention.
In a kind of ad traffic prediction meanss provided in an embodiment of the present invention, can also include:
Update module, for determining the actual advertisement flow of advertisement to be measured, and using actual advertisement flow to inverted index
Table is updated.
The explanation of relevant portion refers to of the invention real in a kind of ad traffic prediction meanss provided in an embodiment of the present invention
The detailed description of corresponding part in a kind of ad traffic prediction technique of example offer is applied, details are not described herein.
The foregoing description of the disclosed embodiments can be realized those skilled in the art or using the present invention.To this
A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can
Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited
It is formed on the embodiments shown herein, and is to fit to consistent with the principles and novel features disclosed in this article widest
Range.
Claims (10)
1. a kind of ad traffic prediction technique characterized by comprising
Obtain the keyword to be measured of advertisement to be measured；
The keyword to be measured is compared with the keyword in inverted index table, is obtained and the Keywords matching to be measured
With reference to keyword, wherein the inverted index table is to build in advance according to the corresponding historical traffic data of advertisement launched in history
Vertical；
It is determining with it is described with reference to the corresponding advertisement of keyword be with reference to advertisement, and by being obtained in the inverted index table and the ginseng
It is corresponding with reference to ad traffic to examine advertisement；
It is calculated using described with reference to ad traffic, obtains prediction ad traffic corresponding with the advertisement to be measured.
2. the method according to claim 1, wherein obtaining the keyword to be measured of the advertisement to be measured, comprising:
Extract keyword corresponding with the advertisement to be measured；
The matched key of stop words in determining keyword corresponding with the advertisement to be measured and in the deactivated vocabulary that pre-establishes
Word is failure keyword；
Failure keyword in keyword corresponding with the advertisement to be measured is deleted, keyword to be measured is obtained.
3. according to the method described in claim 2, it is characterized in that, by the key in the keyword to be measured and inverted index table
Word is compared, and obtains the reference keyword with the Keywords matching to be measured, comprising:
Using binary chop, the keyword to be measured is compared with the keyword in the inverted index table, obtain with
The reference keyword of the Keywords matching to be measured.
4. according to the method described in claim 2, it is characterized in that, calculated using described with reference to ad traffic, obtain with
The corresponding prediction ad traffic of the advertisement to be measured, comprising:
If the quantity with reference to advertisement is one, it is determined that this is described to be measured with reference to the corresponding reference ad traffic of advertisement
The prediction ad traffic of advertisement；If the quantity with reference to advertisement is multiple, it is determined that each described corresponding with reference to advertisement
Average value with reference to ad traffic is the prediction ad traffic of the advertisement to be measured.
5. method according to any one of claims 1 to 4, which is characterized in that in advance according to the advertisement pair launched in history
The process that the historical traffic data answered establishes inverted index table includes:
Obtain historical traffic data, the historical traffic data include the advertisement launched in history, each of launch in history it is wide
Accuse corresponding keyword and ad traffic；
It is described according to the corresponding relationship foundation between advertisement, keyword and the ad traffic for including in the historical traffic data
Arrange concordance list.
6. a kind of ad traffic prediction meanss characterized by comprising
Module is obtained, for obtaining the keyword to be measured of advertisement to be measured；
Matching module, for the keyword to be measured to be compared with the keyword in inverted index table, obtain with it is described to
Survey the reference keyword of Keywords matching, wherein the inverted index table is corresponding according to the advertisement launched in history in advance
What historical traffic data was established；
Determining module, for it is determining with it is described with reference to the corresponding advertisement of keyword be with reference to advertisement, and by the inverted index table
Middle acquisition is corresponding with the reference advertisement to refer to ad traffic；
Computing module obtains predict extensively corresponding with the advertisement to be measured for being calculated using described with reference to ad traffic
Accuse flow.
7. device according to claim 6, which is characterized in that obtaining module includes:
Extraction unit, for extracting keyword corresponding with the advertisement to be measured；
Disabling unit, in determining keyword corresponding with the advertisement to be measured with it is deactivated in the deactivated vocabulary that pre-establishes
The matched keyword of word is failure keyword；
Unit is deleted, for deleting the failure keyword in keyword corresponding with the advertisement to be measured, obtains key to be measured
Word.
8. device according to claim 7, which is characterized in that matching module includes:
Matching unit, for utilizing binary chop, by the keyword in the keyword to be measured and the inverted index table into
Row compares, and obtains the reference keyword with the Keywords matching to be measured.
9. device according to claim 7, which is characterized in that computing module includes:
Computing unit, if being one for the quantity with reference to advertisement, it is determined that this is corresponding with reference to advertisement with reference to advertisement
Flow is the prediction ad traffic of the advertisement to be measured；If the quantity with reference to advertisement is multiple, it is determined that each described
It is the prediction ad traffic of the advertisement to be measured with reference to the corresponding average value with reference to ad traffic of advertisement.
10. according to the described in any item devices of claim 6 to 9, which is characterized in that further include:
Module is established, for obtaining historical traffic data, the historical traffic data includes the advertisement launched in history, in history
The corresponding keyword of each advertisement and ad traffic launched；And according to include in the historical traffic data advertisement, key
Corresponding relationship between word and ad traffic establishes the inverted index table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711200255.2A CN109840784A (en) | 2017-11-24 | 2017-11-24 | A kind of ad traffic prediction technique and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711200255.2A CN109840784A (en) | 2017-11-24 | 2017-11-24 | A kind of ad traffic prediction technique and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109840784A true CN109840784A (en) | 2019-06-04 |
Family
ID=66879699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711200255.2A Pending CN109840784A (en) | 2017-11-24 | 2017-11-24 | A kind of ad traffic prediction technique and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109840784A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465532A (en) * | 2019-09-06 | 2021-03-09 | 百度在线网络技术（北京）有限公司 | Flow estimation method and flow estimation device for information flow advertisement |
-
2017
- 2017-11-24 CN CN201711200255.2A patent/CN109840784A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465532A (en) * | 2019-09-06 | 2021-03-09 | 百度在线网络技术（北京）有限公司 | Flow estimation method and flow estimation device for information flow advertisement |
| CN112465532B (en) * | 2019-09-06 | 2023-10-03 | 百度在线网络技术（北京）有限公司 | Flow estimating method and flow estimating device for information flow advertisement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107066621B (en) | Similar video retrieval method and device and storage medium | |
| CN105654201A (en) | Advertisement flow prediction method and device | |
| CN104281649A (en) | Input method and device and electronic equipment | |
| CN111818313B (en) | Vehicle real-time tracking method and device based on monitoring video | |
| CN105389349A (en) | Dictionary updating method and apparatus | |
| CN103970733B (en) | A kind of Chinese new word identification method based on graph structure | |
| CN110019889A (en) | Training characteristics extract model and calculate the method and relevant apparatus of picture and query word relative coefficient | |
| CN102902826A (en) | Quick image retrieval method based on reference image indexes | |
| CN110070059B (en) | Unstructured road detection method based on domain migration | |
| CN112613569A (en) | Image recognition method, and training method and device of image classification model | |
| CN110990917A (en) | BIM model display method, device and system | |
| CN108509545A (en) | A kind of comment processing method and system of article | |
| CN109840784A (en) | A kind of ad traffic prediction technique and device | |
| CN106445973B (en) | The monitoring method and device of search engine | |
| CN107133274B (en) | Distributed information retrieval set selection method based on graph knowledge base | |
| CN105069663A (en) | Information push method and device based on preference of user | |
| CN110362828A (en) | Network information Risk Identification Method and system | |
| US20230144288A1 (en) | Method for determining intersection missing traffic restriction information, and electronic device | |
| CN103500214B (en) | Word segmentation information pushing method and device based on video searching | |
| CN109799905A (en) | A kind of hand tracking and advertisement machine | |
| CN115391665A (en) | Video recommendation method and device, electronic equipment and storage medium | |
| CN109829048A (en) | Electronic device, interview householder method and computer readable storage medium | |
| CN115878877A (en) | Concept drift-based visual detection method for access crawler of aviation server | |
| CN114136333A (en) | High-precision map road data generation method, device and equipment based on hierarchical features | |
| CN104008146B (en) | Image query expansion method and system based on common visual patterns |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190604 |