Methods, systems, and computer readable media for generating and using a web page classification model

1. A method of generating and using a web page classification model, the method comprising:

identifying a plurality of web pages for use in generating the web page classification model;

assigning a label to each of the plurality of web pages;

accessing, by a processor, Transmission Control Protocol/Internet Protocol (TCP/IP) traffic traces associated with downloading content from each of the plurality of web pages;

processing, by the processor, TCP/IP headers from the TCP/IP traffic traces to identify and extract features that discriminate between the labels, that are uncorrelated and whose discriminatory accuracy remains stable across time and/or browser platform;

generating, by the processor, the web page classification model by training a model trainer to learn a combination of the features that accurately discriminates between the labels, that includes uncorrelated features, and that remains stable across time and/or browser platform; and

using the web page classification model to classify unlabeled web pages by applying the web page classification model to TCP/IP traffic traces used to access the unlabeled web pages.

2. The method of claim 1 , further comprising verifying the accuracy of the web page classification model using observed web page classification data.

3. The method of claim 1 , wherein assigning a label to each of the plurality of web pages comprises assigning an Alexa Genre Label (AGL), a Video Streaming Label (VSL), a Targeted Device Label (TDL), or a Web Page Navigation Label (WNL).

4. The method of claim 3 , wherein assigning the label includes assigning the AGL and wherein the AGL is a content based label indicative of a content genre class of a web page, the content genre class being selected from generic subjects including computers, shopping, business, or news.

5. The method of claim 3 , wherein assigning the label includes assigning the VSL, and wherein the VSL includes at least two classes indicative of the content of a web page, where a first class is video page and a second class is a non-video page.

6. The method of claim 3 , wherein assigning the label includes assigning the TDL and wherein the TDL includes at least two classes indicative of a device used to download a web page, where a first class is a traditional web page and a second class is a mobile optimized web page.

7. The method of claim 3 , wherein assigning the label includes assigning the WNL and wherein the WNL includes at least three classes indicative of a navigation flow used to download a web page, where a first class is a clickable content page, a second class is a search result page, and/or a third class is a landing page.

8. The method of claim 1 , wherein the accessed TCP/IP traces are downloaded using a plurality of different browsers, a plurality of different types of devices, or a plurality of operating systems.

9. The method of claim 1 , wherein processing the TCP/IP headers includes processing anonymized TCP/IP headers.

10. The method of claim 1 , wherein the features include temporal features, multi-flow features, or statistical derivatives thereof.

11. A system for generating a web page classification model, the system comprising:

a processor for accessing Transmission Control Protocol/Internet Protocol (TCP/IP) traffic traces associated with downloading content from a plurality of web pages and processing TCP/IP headers from the TCP/IP traffic traces;

a web page classification model generator executed by the processor for generating the web page classification model, wherein the model generator includes:

a labeler for assigning a label to each of the plurality of web pages;

a feature extractor for identifying and extracting features from the processed TCP/IP headers that discriminate between the labels, that are uncorrelated and whose discriminatory accuracy remains stable across time and/or browser platform; and

a model trainer for generating the web page classification model by learning a combination of the features that accurately discriminates between the labels, that includes uncorrelated features, and that remains stable across time and/or browser platform;

wherein the web page classification model is usable to classify unlabeled web pages by applying the web page classification model to TCP/IP traffic traces used to access the unlabeled web pages.

12. The system of claim 11 , further comprising a model testing and validation module (MTVM) for verifying the accuracy of the web page classification model using observed web page classification data.

13. The system of claim 11 , wherein the label includes at least one of an Alexa Genre Label (AGL), a Video Streaming Label (VSL), a Targeted Device Label (TDL), or a Web Page Navigation Label (WNL).

14. The system of claim 13 , wherein the label includes the AGL and the AGL is a content based label indicative of a content genre class of a web page, the content genre class being selected from generic subjects including computers, shopping, business, or news.

15. The system of claim 13 , wherein the label includes the VSL and wherein the VSL includes at least two classes indicative of the content of a web page, where a first class is video page and a second class is a non-video page.

16. The system of claim 13 , wherein the label includes the TDL and wherein the TDL includes at least two classes indicative of a device used to download a web page, where a first class is a traditional web page and a second class is a mobile optimized web page.

17. The system of claim 13 , wherein the label includes the WNL and wherein the WNL includes at least three classes indicative of a navigation flow used to download a web page, where a first class is a clickable content page, a second class is a search result page, and a third class is a landing page.

18. The system of claim 11 , wherein the TCP/IP traces are downloaded using a plurality of different browsers, a plurality of different types of devices, or a plurality of operating systems.

19. The system of claim 11 , wherein the TCP/IP headers are anonymized.

20. The system of claim 11 , wherein the discriminatory features include temporal features, multi-flow features, or statistical derivatives thereof.

21. A non-transitory computer readable medium having stored thereon executable instructions that when executed by a processor of a computer control the computer to perform steps comprising:

identifying a plurality of web pages for use in generating a web page classification model;

assigning a label to each of the plurality of web pages;

accessing Transmission Control Protocol/Internet Protocol (TCP/IP) traffic traces associated with downloading content from each of the plurality of web pages;

processing TCP/IP headers from the TCP/IP traffic traces to identify and extract features that discriminate between the labels, that are uncorrelated and whose discriminatory accuracy remains stable across time and/or browser platform;

generating the web page classification model by training a classifier to learn a combination of the features that accurately discriminates between the labels, that includes uncorrelated features, and that remains stable across time and/or browser platform; and

using the web page classification model to classify unlabeled web pages by applying the web page classification model to TCP/IP traffic traces used to access the unlabeled web pages.