RELATING WEB PAGE CHANGE WITH REVISITATION PATTERNS

• US 20090327914A1
• Filed: 06/27/2008
• Published: 12/31/2009
• Est. Priority Date: 06/27/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Adar, Eytan, Dumais, Susan T., Liebling, Daniel J., Teevan, Jaime B.
• Granted Patent: US 8,078,974 B2
• Time in Patent Office: Days
• US Class Current: 715/745
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking

Abstract

Web page change may be related to revisitation patterns to support web interaction. In an example embodiment, a method involves analyzing change and revisitation data for a web page, determining a relationship between the data, inferring consumption intent by a user for the web page, and utilizing the inferred consumption intent. More specifically, change data is analyzed to produce a change characterization, with the change data reflecting differences between content of a web page at different times. Revisitation data is analyzed to produce a revisitation characterization, with the revisitation data including visit times to the web page by a user. A relationship is determined between the change and the revisitation data based on the change and the revisitation characterizations. Consumption intent of the user for the content of the web page is inferred responsive to the relationship. The inferred consumption intent is utilized to support interaction with the web page.

Claims

1. One or more processor-accessible tangible media comprising processor-executable instructions for relating web page change with revisitation patterns to support web interaction, wherein the processor-executable instructions, when executed, direct a device to perform acts comprising:

analyzing change data to produce at least one change characterization, the change data reflecting one or more differences between content of a web page at different times, the at least one change characterization comprising a rate, amount, or type of change associated with the web page;

analyzing revisitation data to produce at least one revisitation characterization, the revisitation data including one or more visit times for visits to the web page by at least one user, the at least one revisitation characterization comprising a rate of revisitation corresponding to the web page for the at least one user;

determining a relationship between the change data and the revisitation data based on the at least one change characterization and the at least one revisitation characterization, the relationship between the change data and the revisitation data incorporating a comparison between the rate, amount, or type of change and the rate of revisitation;

inferring a consumption intent of the at least one user for the content of the web page responsive to the relationship between the change data and the revisitation data, the inferring comprising;

inferring that the consumption intent of the at least one user is to consume static content of the web page if the rate, amount, or type of change and the rate of revisitation comport with at least one predetermined criterion;

or

inferring that the consumption intent of the at least one user is to consume dynamic content of the web page if the rate of revisitation and the rate, amount, or type of change do not comport with the at least one predetermined criterion; and

utilizing the inferred consumption intent to support interaction with the web page by presenting the static content or the dynamic content in accordance with the inferred consumption intent.

2. A device-implemented method for relating web page change with revisitation patterns to support web interaction, the method comprising acts of:

analyzing change data to produce at least one change characterization, the change data reflecting one or more differences between content of a web page at different times;

analyzing revisitation data to produce at least one revisitation characterization, the revisitation data including one or more visit times for visits to the web page by at least one user;

determining a relationship between the change data and the revisitation data based on the at least one change characterization and the at least one revisitation characterization;

inferring a consumption intent of the at least one user for the content of the web page responsive to the relationship between the change data and the revisitation data; and

utilizing the inferred consumption intent to support interaction with the web page.

3. The method as recited in claim 2, wherein the act of analyzing change data comprises:

producing at least one change curve that represents the change data, the at least one change curve representing an amount of change to the content of the web page versus time.

4. The method as recited in claim 3, wherein the act of analyzing change data further comprises:

determining a knot point of the at least one change curve, the knot point indicating a time when the amount of change to the content of the web page approximately ceases with respect to reference content from the web page.

5. The method as recited in claim 4, wherein:

the act of analyzing revisitation data comprises producing at least one revisitation curve that represents the revisitation data, the at least one revisitation curve including one or more peaks indicating at least one maximally-frequent inter-visit time; and

the act of determining a relationship comprises comparing (i) the time of the knot point when the amount of change to the content of the web page approximately ceases to (ii) the at least one maximally-frequent inter-visit time of the one or more peaks of the revisitation curve.

6. The method as recited in claim 2, wherein:

the act of analyzing change data comprises producing at least one aggregate change statistic that represents the change data;

the at least one aggregate change statistic comprising an amount of change, an average rate or amount of change, or another summary metric; and

the act of analyzing revisitation data comprises producing at least one aggregate revisitation statistic that represents the revisitation data;

the at least one aggregate revisitation statistic comprising a total number of revisiting users, an average rate of revisitation, an average inter-visit time, or another summary metric.

7. The method as recited in claim 2, wherein the act of utilizing the inferred consumption intent comprises:

presenting at least a portion of the web page responsive to the inferred consumption intent.

8. The method as recited in claim 7, wherein the act of presenting at least a portion of the web page comprises:

presenting current content, old content, or monitored changes to content of the web page responsive to the inferred consumption intent.

9. The method as recited in claim 2, wherein the acts of analyzing change data, analyzing revisitation data, determining a relationship, and inferring a consumption intent comprise:

inputting the change data to a learning machine, the learning machine trained with change information;

inputting the revisitation data to the learning machine, the learning machine trained with revisitation information; and

processing the change data and the revisitation data with the learning machine to output the inferred consumption intent.

10. The method as recited in claim 2, wherein the at least one change characterization comprises a rate of change, and the at least one revisitation characterization comprises a rate of revisitation;

and wherein the act of determining comprises;

determining which of the rate of change or the rate of revisitation is greater.

11. The method as recited in claim 10, wherein the act of inferring comprises:

if the rate of change is determined to be greater than the rate of revisitation, inferring that the at least one user intends to consume static content of the web page; and

if the rate of revisitation is determined to be greater than the rate of change, inferring that the at least one user intends to consume dynamic content of the web page.

12. The method as recited in claim 2, wherein the act of inferring comprises:

inferring whether the at least one user intends to consume dynamic content or static content of the web page responsive to the relationship between the change data and the revisitation data.

13. The method as recited in claim 2, wherein the act of utilizing comprises:

returning to the web page, by a web crawler, at a crawling rate that is determined responsive to the inferred consumption intent.

14. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that a particular portion of the web page is intended to be consumed by at least one user; and

the act of utilizing comprises differentially utilizing the particular portion in generating a snippet or summary for a search result directed to the web page.

15. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that old content of the web page that has been previously viewed is intended to be consumed by the at least one user; and

the act of utilizing comprises presenting to the at least one user the old content of the web page that has been previously viewed.

16. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that monitored changes of content of the web page is intended to be consumed by the at least one user; and

the act of utilizing comprises presenting to the at least one user indications of what content of the web page has changed as part of a search result snippet or summary of the web page.

17. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that a particular portion of the web page is intended to be consumed by the at least one user; and

the act of utilizing comprises emphasizing for presentation the particular portion of the web page while de-emphasizing from presentation at least one other portion of the web page.

18. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that monitored changes of the web page are intended to be consumed by the at least one user; and

the act of utilizing comprises creating an alert for the at least one user when a monitored change is detected.

19. The method as recited in claim 2, wherein:

the act of inferring comprises inferring responsive to the relationship between the change data and the revisitation data that the at least one user intends to consume at least a portion of the web page after the portion changes; and

the act of utilizing comprises pre-fetching the web page after the monitored portion changes and caching the pre-fetched web page so that the at least one user has access to the changed portion of the web page in a disconnected scenario.

20. A device for relating web page change with revisitation patterns to support web interaction, the device comprising:

a change data analyzer to analyze change data to produce at least one change characterization, the change data reflecting one or more differences between content of a web page at different times;

a revisitation data analyzer to analyze revisitation data to produce at least one revisitation characterization, the revisitation data including one or more visit times for visits to the web page by at least one user;

a change and revisitation relationship determiner to determine a relationship between the change data and the revisitation data based on the at least one change characterization and the at least one revisitation characterization;

a consumption intent inference engine to infer a consumption intent of the at least one user for the content of the web page responsive to the relationship between the change data and the revisitation data; and

a consumption intent utilizer to utilize the inferred consumption intent to support interaction with the web page.