Identifying web pages that are likely to guide browsing viewers to improve conversion rate

• US 9,104,767 B2
• Filed: 08/28/2012
• Issued: 08/11/2015
• Est. Priority Date: 08/28/2012
• Status: Active Grant

• Current Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Suh, Bongwon, Koh, Eunyee
• Primary Examiner(s): RIES, LAURIE ANNE
• Application Number:
  US13/597,155
• Publication Number:
  US 20140068407A1
• Time in Patent Office: 1,078 Days
• Field of Search: 715/205, 715/208, 715/234
• US Class Current: 1/1
• CPC Class Codes: G06F 16/94 Hypermedia Hyperlinking G06F40/134
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A method and system for extracting from a web log for a given entity information comprising web page events. The information gain in the sequence of navigating from a given web page to a subsequent web page is computed for at least some of the web page events. The information gain is used to determine a set of web pages from which users can decide to navigate to a first web page or a second web page, wherein navigating to the first web page is more likely to lead to a conversion than navigating to the second web page. A list of at least a subset of the set of web pages is transmitted to the entity, which can use the list to determine web pages at which the entity can place advertising material to motivate a user to navigate to the first web page. The computing may be performed by MapReduce processing.

Claims

1. A method comprising:

using at least one computer processor, extracting from a web log information comprising web-page events;

identifying from the web-page events a web page that is included in a trail of web pages, which guide browsing sessions to conversions on a plurality of browsed-to web pages that are browsed to after the web page;

calculating a first entropy that quantifies an amount of conversion uncertainty in the trail of web pages before the one or more browsing sessions navigate to the web page;

calculating a second entropy that quantifies an amount of conversion uncertainty in the trail of web pages after the one or more browsing sessions navigate to the web page; and

computing an information gain from the first entropy to the second entropy, wherein the information gain suggests whether a significant conversion disparity exists among the plurality of browsed-to web pages.

2. The method of claim 1 further comprising transmitting a list of at least the web page for displaying to an entity.

3. The method of claim 1 wherein the conversions are customizable.

4. The method of claim 1 wherein the conversions comprise at least one of the group consisting of downloading a file, adding a product to a cart, sending credit card information, and purchasing the product.

5. The method of claim 1 wherein information extracted from the web-page events includes at least some of the group consisting of timestamps, LTRLs, credit card information, and product purchase information.

6. The method of claim 2 wherein the list indicates at least the web page at which the entity can place advertising material to motivate a user to navigate to one or more webpages of the plurality of browsed-to web pages that are browsed to after the web page.

7. The method of claim 1 wherein the computing comprises processing URL pairs that comprise a first URL from which users navigate and a second URL to which users navigate from the first URL.

8. The method of claim 7 wherein the computing comprises processing a MapReduce function.

9. A machine-readable storage device having stored therein a set of instructions which, when executed by the machine, causes the machine to execute the following operations:

extracting from a web log information comprising web-page events;

identifying from the web-page events a web page that is included in a trail of web pages, which guide browsing sessions to conversions on a plurality of browsed-to web pages that are browsed to after the web page;

calculating a first entropy that quantifies an amount of conversion uncertainty in the trail of web pages before the one or more browsing sessions navigate to the web page;

calculating a second entropy that quantifies an amount of conversion uncertainty in the trail of web pages after the one or more browsing sessions navigate to the web page; and

computing an information gain from the first entropy to the second entropy, wherein the information gain suggests whether a significant conversion disparity exists among the plurality of browsed-to web pages.

10. The machine-readable storage device of claim 9 further comprising transmitting a list of at least the web page for displaying to an entity.

11. The machine-readable storage device of claim 9 wherein the conversions are customizable.

12. The machine-readable storage device of claim 9 wherein the conversions comprise at least one of the group consisting of downloading a file, adding a product to a cart, sending credit card information, and purchasing the product.

13. The machine-readable storage device of claim 9 wherein information extracted from the web-page events includes at least some of the group consisting of timestamps, URLs, credit card information, and product purchase information.

14. The machine-readable storage device of claim 10 wherein the indicates at least the web page at which the entity can place advertising material to motivate a user to navigate to one or more webpages of the plurality of browsed-to web pages that are browsed to after the web page.

15. The machine-readable storage device of claim 9 wherein the computing comprises processing URL pairs that comprise a first URL from which users navigate and a second URL to which users navigate from the first URL.

16. The machine-readable storage device of claim 15 wherein the computing comprises processing a MapReduce function.

17. A system comprising:

at least one computer processor configured to

extract from a web log information comprising web-page events;

identify from the web-page events a web page that is included in a trail of web pages, which guide browsing sessions to conversions on a plurality of browsed-to web pages that are browsed to after the web page;

calculate a first entropy that quantifies an amount of conversion uncertainty in the trail of web pages before the one or more browsing sessions navigate to the web page;

calculate a second entropy that quantifies an amount of conversion uncertainty in the trail of web pages after the one or more browsing sessions navigate to the web page; and

compute an information gain from the first entropy to the second entropy, wherein the information gain suggests whether a significant conversion disparity exists among the plurality of browsed-to web pages.

18. The system of claim 17, the at least one computer processor further configured to transmit a list of at least the web page for display to an entity.

19. The system of claim 17 wherein conversions are customizable.

20. The system of claim 17 wherein the conversions comprise at least one of the group consisting of downloading a file, adding a product to a cart, sending credit card information, and purchasing the product.

21. The system of claim 17 wherein information extracted from the web-page events includes at least some of the group consisting of timestamps, URLs, credit card information, and product purchase information.

22. The system of claim 17 wherein the computing comprises processing URL pairs that comprise a first URL from which users navigate and a second URL to which users navigate from the first URL.

23. The system of claim 22 wherein the computing comprises processing a MapReduce function.