SYSTEM AND METHOD FOR PROVIDING ADVERTISING SERVER OPTIMIZATION FOR ONLINE COMPUTER USERS

• US 20100250361A1
• Filed: 12/22/2009
• Published: 09/30/2010
• Est. Priority Date: 03/30/2009
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Torigoe, Kendra, Desai, Vinay, Dey, Santanu
• Granted Patent: US 9,547,865 B2
• Time in Patent Office: Days
• US Class Current: 705/14.420
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0243 Comparative campaigns

Abstract

A system and method for providing advertising server optimization for online computer users is disclosed. A particular embodiment includes obtaining an advertisement; defining a set of features of the advertisement; generating a predicted click through rate (CTR) corresponding to the advertisement based in part on the set of features; ranking the advertisement based on the generated click through rate; and serving the advertisement based on the ranking.

Claims

1. A method, including:

obtaining an advertisement;

defining a set of features of the advertisement;

generating a predicted click through rate (CTR) corresponding to the advertisement based in part on the set of features;

ranking the advertisement based on the generated click through rate; and

serving the advertisement based on the ranking.

2. The method as claimed in claim 1 wherein defining a set of features includes defining a plurality of variables associated with each feature of the set of features.

3. The method as claimed in claim 2 wherein defining a set of features includes obtaining values for the plurality of variables associated with each feature of the set of features.

4. The method as claimed in claim 2 including determining a refresh interval corresponding to each of the plurality of variables associated with each feature of the set of features.

5. The method as claimed in claim 1 wherein the set of features includes at least one feature type from the group:

tokenized features, term frequency-inverse document frequency (TF-IDF) features, meta features, lexical features, interaction features, and lookup features.

6. The method as claimed in claim 2 wherein the plurality of variables includes both real-time variables and offline variables.

7. The method as claimed in claim 2 including applying coefficients to each of the plurality of variables in an optimization model.

8. The method as claimed in claim 7 including using the optimization model to generate the predicted click through rate (CTR) corresponding to the advertisement.

9. The method as claimed in claim 1 including determining if an advertising campaign conforms to a pre-defined budget.

10. A system, comprising:

a processor;

an advertising optimization module being executed by the processor to obtain an advertisement, define a set of features of the advertisement, generate a predicted click through rate (CTR) corresponding to the advertisement based in part on the set of features, and rank the advertisement based on the generated click through rate; and

an advertising server, coupled to the advertising optimization module, to serve the advertisement based on the ranking.

11. The system as claimed in claim 10 being further configured to define a plurality of variables associated with each feature of the set of features.

12. The system as claimed in claim 11 being further configured to obtain values for the plurality of variables associated with each feature of the set of features.

13. The system as claimed in claim 11 being further configured to determine a refresh interval corresponding to each of the plurality of variables associated with each feature of the set of features.

14. The system as claimed in claim 10 wherein the set of features includes at least one feature type from the group:

tokenized features, term frequency-inverse document frequency (TF-IDF) features, meta features, lexical features, interaction features, and lookup features.

15. The system as claimed in claim 11 wherein the plurality of variables includes both real-time variables and offline variables.

16. The system as claimed in claim 11 being further configured to apply coefficients to each of the plurality of variables in an optimization model.

17. The system as claimed in claim 16 being further configured to use the optimization model to generate the predicted click through rate (CTR) corresponding to the advertisement.

18. The system as claimed in claim 10 being further configured to determine if an advertising campaign conforms to a pre-defined budget.

19. A machine-useable storage medium embodying instructions which, when executed by a machine, cause the machine to:

obtain an advertisement;

define a set of features of the advertisement;

generate a predicted click through rate (CTR) corresponding to the advertisement based in part on the set of features;

rank the advertisement based on the generated click through rate; and

serve the advertisement based on the ranking.

20. The machine-useable storage medium as claimed in claim 19 being further configured to define a plurality of variables associated with each feature of the set of features.

21. The machine-useable storage medium as claimed in claim 20 being further configured to obtain values for the plurality of variables associated with each feature of the set of features.

22. The machine-useable storage medium as claimed in claim 20 being further configured to determine a refresh interval corresponding to each of the plurality of variables associated with each feature of the set of features.

23. The machine-useable storage medium as claimed in claim 19 wherein the set of features includes at least one feature type from the group:

tokenized features, term frequency-inverse document frequency (TF-IDF) features, meta features, lexical features, interaction features, and lookup features.