Context transfer in search advertising

1. A computer-implemented method for determining a type of landing page to which to transfer web searchers that enter a particular query, each query related to a landing page that links from an advertisement, the method comprising:

extracting, by a computer, content of each of a plurality of landing pages to be classified by a web crawler of a classifier, the content comprising words of the landing page;

extracting, by a computer, text from the content by a page rendering program of the classifier;

establishing, by a computer, a feature space based on the extracted text, representing the extracted landing page;

reducing, by a computer, the feature space by applying a supervised attribute selection technique to the classifier; and

training, by a computer, a machine learning model of the classifier using a learning algorithm;

classifying, by a computer, a target landing page as one of a plurality of landing page classes with the classifier trained by the machine learning model;

determining, by a computer, characteristics of one or more query;

associating the target landing page with the one or more query based on the class of the landing page being related to the characteristics of the one or more query; and

choosing, by a computer, whether to retain or to change classification of the target landing page associated with the one or more query based on relative average conversion rates of advertisements on a plurality of previously-classified landing pages when associated with the characteristics of the one or more query,

wherein the previously-classified landing pages are different than the landing page, and

wherein relative average conversion rates comprise average conversion rates that result for respective classified type of landing page over a period of time.

2. The method of claim 1 , wherein the classes comprises at least one of homepage, search transfer, category browse, miscellaneous, or another category.

3. The method of claim 1 , wherein the query characteristics comprise one or more of query frequency, query length, and query topic, wherein the query topic comprises a query class predicted by an automatic query classifier with respect to a commercial taxonomy.

4. The method of claim 1 , wherein the classification by the classifier is based on feature extraction comprising:

observing particular characteristics on each landing page to distinguish among the plurality of landing page classes; and

detecting particular words within the text of each landing page to be correlated with the plurality of landing page classes.

5. The method of claim 1 , wherein the content extracted by the web crawler is hypertext markup language (HTML) content.

6. A computer-implemented method for determining a type of landing page to which to transfer web searchers that enter a particular query, the queries related to landing pages that link from respective advertisements, the method comprising:

extracting, by a computer, content of each of a plurality of landing pages to be classified by a web crawler of a classifier, the content comprising words of the landing page;

extracting, by a computer, text from the content by a page rendering program of the classifier;

establishing, by a computer, a feature space based on the extracted text, representing the extracted landing page;

reducing, by a computer, the feature space by applying a supervised attribute selection technique to the classifier; and

training, by a computer, a machine learning model of the classifier using a learning algorithm;

classifying, by a computer, each of a plurality of target landing pages into one of a plurality of classes with the classifier trained by the machine learning model using content extracted from each of the plurality of target landing pages, wherein the plurality of classes comprise homepage, search transfer, and category browse;

gathering, by a computer, characteristics on one or more query associated with each target landing page;

partitioning, by a computer, the plurality of target landing pages according to a plurality of query characteristics;

determining, by a computer, conversion rates of advertisements on at least some of the plurality of target landing pages;

correlating, by a computer, the plurality of classes of landing pages within each landing page partition with corresponding conversion rates; and

choosing a target landing page of a particular class to associate with an identified query within a search engine based on the corresponding conversion rate of that target landing page class as associated with the characteristics of the identified query.

7. The method of claim 6 , wherein the classes further comprise miscellaneous, wherein miscellaneous includes landing pages not falling into a classification of homepage, search transfer, or category browse.

8. The method of claim 6 , wherein the plurality of query characteristics comprise one or more of query frequency, query length, and query topic.

9. The method of claim 8 , wherein the query topic comprises a query class predicted by an automatic query classifier with respect to a commercial taxonomy.

10. The method of claim 6 , wherein the content extracted by the web crawler is hypertext markup language (HTML) content.

11. The method of claim 10 , wherein the classification by the classifier is based on feature extraction comprising:

observing particular characteristics on the landing pages to distinguish among the plurality of landing page classes; and

detecting particular words within the text of each landing page to be correlated with the plurality of landing page classes.

12. The method of claim 10 , wherein the machine learning model comprises a support vector machine model and wherein the learning algorithm comprises a sequential minimal optimization (SMO) algorithm of the support vector machine model.

13. A server for determining a type of landing page to which to transfer web searchers that enter a particular query, each query related to a landing page that links from an advertisement, the server comprising:

a landing page classifier to

extract content of each of a plurality of landing pages to be classified, the content comprising words of the landing page;

extract text from the content;

establish a feature space based on the extracted text, representing the extracted landing page;

reduce the feature space by applying a supervised attribute selection technique; and

conduct classification training with a computer under a machine learning model using a learning algorithm; and

classify a target landing page as one of a plurality of landing page classes;

a query classifier coupled with the landing page classifier to determine characteristics of one or more query such that the one or more query is associated with the target landing page based on the class of the landing page being related to the characteristics of the one or more query; and

a processor coupled with the landing page and query classifiers to choose whether to retain or to change classification of the target landing page associated with the one or more query based on relative average conversion rates of advertisements on a plurality of previously-classified landing pages when associated with the characteristics of the one or more query,

wherein the previously-classified landing pages are different than the landing page, and

wherein relative average conversion rates comprise average conversion rates that result for respective classified type of landing page over a period of time.

14. The server of claim 13 , wherein the classes comprises at least one of homepage, search transfer, category browse, miscellaneous, or another category,

wherein the query characteristics comprise one or more of query frequency, query length, and query topic, and

wherein the query topic comprises a query class predicted by an automatic query classifier with respect to a commercial taxonomy.

15. The server of claim 13 , wherein the processor observes particular characteristics on the target landing page to distinguish among the plurality of landing page classes,

the server further comprising:

an extractor to extract particular words within the text of each target landing page to be correlated as features with the plurality of landing page classes.

16. The server of claim 13 , wherein the classifier comprises a web crawler to extract hypertext markup language (HTML) content and a page rendering program to extract text from the HTML content of each landing page to be classified, wherein the machine learning model comprises a support vector machine model, and

wherein the learning algorithm comprises a sequential minimal optimization (SMO) algorithm of the support vector machine model.

17. A computer-implemented system for determining a type of landing page to which to transfer web searchers that enter a particular query, the queries related to landing pages that link from respective advertisements, the system comprising:

a landing page classifier to

extract content of each of a plurality of landing pages to be classified, the content comprising words of the landing page;

extract text from the content;

establish a feature space based on the extracted text, representing the extracted landing page;

reduce the feature space by applying a supervised attribute selection technique; and

conduct classification training with a computer under a machine learning model using a learning algorithm; and

classify each of a plurality of target landing pages into one of a plurality of classes using content extracted from each of the target plurality of landing pages, the content comprising words of the landing page, wherein the plurality of classes comprise homepage, search transfer, and category browse;

a processor coupled with a memory and with a communication interface;

a query logs database coupled with the processor, wherein the processor is configured to:

gather from the query logs database characteristics on one or more query associated with each target landing page;

partition the target landing pages according to a plurality of query characteristics;

retrieve conversion rates of advertisements on at least some of the plurality of target landing pages;

correlate the plurality of classes of landing pages within each target landing page partition with corresponding conversion rates; and

choose a target landing page of a particular class to associate with an identified query within a search engine based on the corresponding conversion rate of that landing page class as associated with the characteristics of the identified query.

18. The system of claim 17 , wherein the classes further comprise miscellaneous, or another category, and wherein the query characteristics comprise one or more of query frequency, query length, and query topic.

19. The system of claim 18 , wherein the query topic comprises a query class predicted by an automatic query classifier with respect to a commercial taxonomy.

20. The server of claim 17 , wherein the classifier comprises a web crawler to extract hypertext markup language (HTML) content and a page rendering program to extract text from the HTML content of each landing page to be classified, wherein the processor observes particular characteristics on each landing page to distinguish among the plurality of landing page classes, the server further comprising:

an extractor to extract particular words within the text of each landing page to be correlated as features with the plurality of landing page classes.