Personalizing Sponsored Search Advertising Layout using User Behavior History

• US 20090240677A1
• Filed: 03/18/2008
• Published: 09/24/2009
• Est. Priority Date: 03/18/2008
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Parekh, Rajesh, Berkhin, Pavel, Parmar, Jignesh
• Granted Patent: US 8,762,364 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Embodiments of the invention relate to methods of presenting personalized search results pages to users, and to search engine systems and servers configured to implement such methods. For example, a method of presenting such a page to a user of a search engine includes steps of computing an engagement index of the user based on the distribution in time of that user'"'"'s interactions with the search engine then presenting, in response to a query by the user, a personalized search results page to the user.

Claims

1. A computer-implemented method of presenting a personalized search results page to a user of a search engine, comprising:

a. computing an engagement index of a user based on the distribution in time of that user'"'"'s interactions with the search engine;

b. presenting, in response to a query by the user, a personalized search results page to the user if the engagement index of the user meets a threshold.

2. The computer-implemented method of claim 1 where the engagement index is additionally based user interest data inferred from the distribution of the users interactions with non-search properties.

3. The computer-implemented method of claim 2, wherein the non-search properties include one of the following:

autos, finance, and sports.

4. A web search server embodied in a computer-readable medium and configured to personalize layout of a web search results page of a search engine, comprising:

a. a personalization module configured to receive user behavior data concerning a user'"'"'s interactions with a search engine and to determine an engagement index for the user based on some of the user behavior data, and to generate, in response to a query by the user, a search results page with an advertising layout;

i. based on the user behavior data when the engagement index for the user meets an engagement threshold; and

ii. not based on the user behavior data when the engagement index for the user does not meet the engagement threshold.

5. The web search server of claim 4, wherein the advertising layout based on the user behavior data is expected to generate additional revenue relative to the advertising layout not based on the user behavior data.

6. The web search server of claim 4, wherein the advertising layout based on the user behavior data decreases the number of ads shown in a region of the search results page relative to the advertising layout not based on the user behavior data.

7. The web search server of claim 4, configured to serve a first user and a second user wherein an advertising layout based on first user behavior data is expected to generate additional revenue relative to the advertising layout not based on the user behavior data and wherein an advertising layout based on second user behavior data decreases the number of ads shown in a region of the search results page relative to the advertising layout not based on the user behavior data.

8. The web search server of claim 7, wherein the first user is classified as a heavy clicker and the second user is classified as a light clicker.

9. The web search server of claim 4, wherein the aspect of the advertising layout that is modified is the number of ads shown in each of the of the possible ad locations on the search results page.

10. A computer-implemented method of personalizing layout of a web search results page of a search engine, comprising:

a. collecting user behavior data concerning a user'"'"'s interactions with a search engine;

b. determining an engagement index for the user based on some of the user behavior data; and

c. presenting to the user, in response to a query by the user, a search results page with an advertising layout;

i. based on the user behavior data when the engagement index for the user meets an engagement threshold; and

ii. not based on the user behavior data when the engagement index for the user does not meet the engagement threshold.

11. The method of claim 10, wherein the advertising layout based on the user behavior data is expected to generate additional revenue relative to the advertising layout not based on the user behavior data.

12. The method of claim 10, wherein the advertising layout based on the user behavior data decreases the number of ads shown in a region of the search results page relative to the advertising layout not based on the user behavior data.

13. The method of claim 10, wherein the aspect of the advertising layout that is modified is the number of ads shown in each of the of the possible ad locations on the search results page.

14. The method of claim 10, where the aspect of the advertising layout that is modified is the number of ads shown in a primary advertising location of the search results page.

15. The method of claim 14, wherein the primary advertising location is the north advertising section of the search results page.

16. The method of claim 10, wherein the user behavior data comprises a number of pages returned to the user'"'"'s browser and a number of clicks the user has performed on sponsored listings, in combination with the content and location of the sponsored listings.

17. The method of claim 10, wherein if the engagement index for the user meets the engagement threshold the user is classified based on user behavior data as one of the following:

heavy clicker, average clicker, and light clicker.

18. The method of claim 17, wherein same-classified users are presented with the same advertising layout.

19. The method of claim 10, wherein the advertising layout presented to a user whose engagement index meets the engagement threshold is based on the user'"'"'s historical click-through rate (CTR).

20. The method of claim 19, wherein the historical user click-through rate (CTR) is used as a scaling factor to increase, decrease, or preserve the expected revenue on an advertisement given the query and the user.

21. The method of claim 19, wherein the historical user click-through rate (CTR) is used as a factor in modeling the expected revenue on an advertisement given a query and the user along with other factors derived from the query and the advertisement.