Improving results from search providers using a browsing-time relevancy factor

• US 8,635,214 B2
• Filed: 07/26/2006
• Issued: 01/21/2014
• Est. Priority Date: 07/26/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Blass, Oscar J., Gago, Oswaldo, Monteiro, Brennan D., Patel, Paritosh D., Vila, Roberto
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): VO, CECILE H
• Application Number:
  US11/460,038
• Publication Number:
  US 20080059446A1
• Time in Patent Office: 2,736 Days
• Field of Search: 707/3, 707/731, 375148-149, 375/354
• US Class Current: 707/731
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A method for searching Web pages that begins with the identification of query criteria entered into a search provider. A set of Web pages that satisfies the query criteria are determined. Then, a page ranking is ascertained for each Web page in the set. The Web pages are presented in order by page ranking. The page ranking is based upon at least one relevancy factor that includes a browsing-time factor. The browsing-time factor can be calculated from browsing behavior exhibited by users, who provided similar query criteria. The set of users from which the browsing-time factor is calculated can include a current user, a set of users sharing characteristics with the current user, and/or a general set of users. Browsing behavior can include time spent at a Web page, where the browsed Web page is a page that was previously presented as a search result for the similar query criteria.

Claims

1. A method for searching Web pages comprising:

identifying query criteria entered into a search provider;

determining a plurality of Web pages that satisfy the query criteria;

ascertaining a page ranking for each of the plurality of Web pages, wherein each page ranking is based upon at least one relevancy factor; and

presenting ordered results for the query criteria, which are ordered by the ascertained page rankings, wherein the at least one relevancy factor includes a browsing-time factor, wherein the browsing-time factor is based in part upon a time spent at a browsed Web page, and wherein the browsing-time factor is adjusted responsive to an adjustment event, wherein the adjustment event indicates a level of user interest in the browsed Web page based on previous browsing of the web page, wherein the level of user interest is determined without regard to time spent at the browsed Web page, wherein the browsing-time factor is adjusted by detecting a positive adjustment event relating to the browsed Web page, and favorably adjusting the browsing-time factor responsive to the positive adjustment event, wherein the positive adjustment event indicates a user interest in the browsed Web page.

2. The method of claim 1, further comprising:

using client-side software to monitor time spent at the browsed Web page; and

conveying data gathered by the client-side software to the search provider, said search provider using the conveyed data to quantify the browsing-time factor.

3. The method of claim 1, wherein a user enters the query criteria into a search Web page, wherein the presenting of results occurs via the search Web page, and wherein the browsing-time factor is quantified based upon a time it takes for the user to return to the search Web page after having selected the browsed Web page from the results presented within the search Web page.

4. The method of claim 1, wherein the time spent at the browsed Web page is an average time spent by a plurality of different users at the browsed Web page responsive to a search based upon criteria similar to the query criteria.

5. The method of claim 4, further comprising:

limiting an affect that each of the different users has upon the browsing-time factor compared to the other ones of the different users so that no one user is able to dominate a calculation of the time spent at the browsed Web page.

6. The method of claim 1, wherein the positive adjustment event comprises at least one event selected from a group consisting of a user viewing the browsed Web page in its entirety, a user downloading a file from the browsed Web page, a user printing at least a portion of the browsed Web page, a user making a purchase from the browsed Web page, a user adding the browsed Web page to a list of browser favorites, a user not accessing other Web pages included in the ordered results after the browsed Web page has been accessed, and a user entering dissimilar criteria from the query criteria a next time the search provider is used by the user.

7. The method of claim 1, wherein said steps of claim 1 are steps performed by at least one of a service agent and a computing device manipulated by the service agent, the steps being performed in response to a service request.

8. A method for ordering a set of Web pages satisfying query criteria comprising:

for each Web page in a set of Web pages satisfying query criteria, assigning a page ranking value based at least in part upon a browsing-time factor;

ordering the Web pages in the set in accordance with the assigned page ranking values; and

presenting the Web pages in order within a search Web page in which the query criteria was entered, wherein the browsing-time factor is based at least in part upon a time spent by at least one user at an associated browsed Web page, which was browsed responsive to a user selection of a hyperlink presented within the search Web page, said hyperlink being presented as part of a result set from at least one previous query having criteria similar to the query criteria, and wherein the browsing time factor is adjusted responsive to an adjustment event, wherein the adjustment event indicates a level of user interest in the browsed Web page as part of a result set from at least one previous query having criteria similar to the query criteria, wherein the level of user interest is determined without regard to time spent at the browsed Web page, wherein the browsing-time factor is adjusted by detecting a negative adjustment event relating to the browsed Web page, and unfavorably adjusting the browsing-time factor responsive to the negative adjustment event, wherein the negative adjustment event indicates a user disinterest in the browsed Web page.

9. The method of claim 8, wherein the time spent at the browsed Web page is an average time spent by a plurality of different users at the browsed Web page response to a search based upon criteria similar to the query criteria.

10. The method of claim 9, further comprising:

limiting an affect that each of the different users has upon the browsing-time factor compared to the other ones of the different users so that no one user is able to dominate a calculation of the time spent at the browsed Web page.

11. The method of claim 8, wherein the negative adjustment event comprises at least one event selected from a group consisting of a user not viewing the entirety of the browsed Web page, a user adding the browsed Web page to a list of Web pages blocked by a firewall, a user accessing other ones of the Web pages included in the ordered results, and a user entering similar criteria to the query criteria a next time the search provider is used by the user.

12. A system for ordering a set of Web pages comprising:

a computer executing a search provider configured to search Web pages for a user-entered query criteria; and

a computer executing a result refinement engine configured to apply relevancy algorithms, wherein said relevancy algorithms order Web pages based on at least one relevancy factor, wherein the at least one relevancy factor includes a browsing-time factor, wherein the browsing-time factor is based in part upon a time spent at a browsed Web page, and wherein the browsing time factor is adjusted responsive to an adjustment event, wherein the adjustment event indicates a level of user interest in the browsed Web page based on previous browsing of the web page, wherein the level of user interest is determined without regard to time spent at the browsed Web page, wherein the browsing time factor is adjusted by detecting a negative adjustment event relating to the browsed Web page, and unfavorably adjusting the browsing-time factor responsive to the negative adjustment event, wherein the negative adjustment event indicates a user disinterest in the browsed Web page.

13. The system of claim 12, further comprising:

at least one client, wherein the at least one client includes a behavior capture engine configured to capture data used to determine the browsing-time factor of the browsed Web page.

14. The system of claim 12, further comprising:

a server configured to serve a set of Web pages containing content derived from results of said search provider, wherein the server includes a return-to-engine timer configured to capture data used to determine the browsing-time factor of the browsed Web page.