Method and system for quantifying user interactions with web advertisements
Methods and systems are provided that may be used to determine a probability of whether a visitor to a web document is likely to click on a web advertisement. An exemplary method may include detecting one or more features in a web document. One or more expert statistical models to which the web document belongs may be determined and associated weightings may be determined based, at least in part, on the one or more features detected. A click-through-rate probability for a web advertisement to be placed on the web document may be estimated based on the one or more expert statistical models.
1. A method comprising:
determining, by one or more processors, one or more first features of a web document;
determining, by the one or more processors, relevance scores for individual ones of a set of two or more pre-existing expert statistical models based at least in part on comparison between the one or more first features of the web document and one or more second features of the set of two or more pre-existing expert statistical models, wherein at least one of the individual ones of the set of two or more pre-existing expert statistical models comprises a statistical model to assess at least the one or more first features of the web document;
selecting, by the one or more processors, one or more expert statistical models, from the set of two or more pre-existing expert statistical models based, at least in part, on the relevance scores;
determining, by the one or more processors, weightings for the one or more selected expert statistical models based, at least in part, on the relevance scores for the one or more expert statistical models;
assessing, by the one or more processors, the at least the one or more first features of the web document based, at least in part, on the one or more selected expert statistical models; and
estimating, by the one or more processors, a click-through-rate probability for a web advertisement to be placed on the web document based on the weightings for the one or more selected expert statistical models.
2. The method of claim 1 , wherein the web document comprises at least one of a web page or a search query.
3. The method of claim 1 , wherein the relevance score between the web document and a particular expert statistical model is based, at least in part, on the one or more first features selected.
4. The method of claim 1 , further comprising determining expected revenue for the web advertisement based at least in part on the click-through-rate probability and an auction bid corresponding to the web advertisement.
5. The method of claim 1 , wherein the estimating the click-through-rate probability is based, at least in part, on at least one of a publisher identifier corresponding to a publisher of the web document, a predetermined position in which the publisher locates web advertisements on the web document, or a relevance score based on a comparison of the one or more first features in the web document and one or more third features of the web advertisement.
6. The method of claim 1 , wherein the one or more first features comprise at least one of key words located in the web document, phrases detected in the web document, or taxonomy of content in the web document.
7. The method of claim 1 , wherein the determining weightings for one or more expert statistical models comprises assigning weightings to the one or more first features detected in the web document, and selecting the one or more expert statistical models based at least in part on combined weightings of the one or more first features.
8. An article comprising:
a non-transitory storage medium comprising machine-readable instructions stored thereon which are executable by one or more processors to:
detect one or more first features of a web document;
determine relevance scores for individual ones of a set of two or more pre-existing expert statistical models based at least in part on comparison between the one or more first features of the web document and one or more second features of the set of two or more pre-existing expert statistical models, wherein at least one of the individual ones of the set of two or more pre-existing expert statistical models comprises a statistical model to assess at least the one or more first features of the web document;
select one or more expert statistical models, from the set of two or more pre-existing expert statistical models based, at least in part, on the relevance scores;
determine weightings for one or more selected expert statistical models based, at least in part, on the one or more relevance scores for the expert statistical models;
assess the at least the one or more first features of the web document based, at least in part, on the one or more selected expert statistical models; and
estimate, based on the weightings for the one or more selected expert statistical models, a click-through-rate probability for a web advertisement to be placed on the web document.
9. The article of claim 8 , wherein the machine-readable instructions are further executable by the one or more processors to determine the set of two or more expert statistical models to predict the click-through-rate probability for web documents and web advertisements based on an analysis of the one or more features in the web documents and the web advertisements.
10. The article of claim 8 , wherein the machine-readable instructions are further executable by the one or more processors to select the one or more expert statistical models from the set of two or more expert statistical models based, at least in part, on the one or more features selected.
11. The article of claim 8 , wherein the machine-readable instructions are further executable by the one or more processors to estimate the click-through-rate probability based, at least in part, on at least one of a publisher identifier corresponding to a publisher of the web document, a predetermined position in which the publisher locates one or more web advertisements on the web document, or a relevance score based on a comparison of the one or more features in the web document and one or more features of the web advertisement.
12. The article of claim 8 , wherein the machine-readable instructions are further executable by the one or more processors to assign weightings to one or more features detected in the web document, and select and determined the weightings for the one or more expert statistical models based on combined weightings of the one or more features.
13. A system comprising:
a computing platform comprising one or more processors to:
detect one or more first features of a web document;
determine relevance scores for individual ones of a set of two or more pre-existing expert statistical models based at least in part on comparison between the one or more first features of the web document and one or more second features of the set of two or more pre-existing expert statistical models, wherein at least one of the individual ones of the set of two or more pre-existing expert statistical models comprises a statistical model to assess at least the one or more first features of the web document;
select one or more expert statistical models, from the set of two or more pre-existing expert statistical models based, at least in part, on the relevance scores;
determine weightings for one or more selected expert statistical models based, at least in part, on the one or more relevance scores for the expert statistical models;
assess the at least the one or more first features of the web document based, at least in part, on the one or more selected expert statistical models; and
estimate, based on the weightings for the one or more selected expert statistical models, a click-through-rate probability for a web advertisement to be placed on the web document.
14. The system of claim 13 , wherein the web document comprises at least one of a web page or a search query.
15. The system of claim 13 , wherein the computing platform is capable of determining the set of two or more expert statistical models to predict the click-through-rate probability for web documents and web advertisements based on an analysis of the one or more first features in the web documents and the web advertisements.
16. The system of claim 13 , wherein computing platform is capable of determining expected revenue for the web advertisement based on the click-through-rate probability and an auction bid corresponding to the web advertisement.
17. The system of claim 13 , wherein the computing platform is capable of estimating the click-through-rate probability based, at least in part, on at least one of a publisher identifier corresponding to a publisher of the web document, a predetermined position in which the publisher locates web advertisements on the web document, or a relevance score based on a comparison of the one or more features in the web document and one or more features of the web advertisement.
18. The system of claim 13 , wherein the one or more first features comprise at least one of key words located in the web document, phrases detected in the web document, and taxonomy of content in the web document.