Method and system for providing targeted documents based on concepts automatically identified therein

• US 8,520,982 B2
• Filed: 10/18/2011
• Issued: 08/27/2013
• Est. Priority Date: 05/10/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Chan, Wesley T., Agarwal, Sumit, Ranganath, Rama
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US13/276,008
• Publication Number:
  US 20120087591A1
• Time in Patent Office: 679 Days
• Field of Search: 382/305, 382/181, 382/218, 707/736, 705/1, 705/4, 705/14.1, 705 144- 1443, 705/7.29, 713/3, 713/4, 713/6
• US Class Current: 382/305
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0263 based upon Internet or website rating

Abstract

A system and method for providing a graphical document are provided. A graphical document is processed to identify one or more ideas associated with the graphical document. The identified ideas may be used to determine concepts to associate with the first electronic document and suggest those concepts for association with the first electronic document for use in a document distribution system that distributes documents based on a price parameter associated with concepts and a performance parameter. A request for a document associated with a concept is received. Responsive to the request, the graphical document is delivered if an association between the one or more ideas and the concept is determined.

Claims

1. A computer-implemented method comprising:

comparing image data that specify optical content depicted by a first advertisement to image data that specify optical content depicted by a reference advertisement, wherein the reference advertisement is associated with at least one keyword;

determining, by one or more processors, a similarity measure between the first advertisement and the reference advertisement based on the comparison of the image data;

associating one or more of the keywords from the reference advertisement with the first advertisement based on the similarity measure between the first advertisement and the reference advertisement;

receiving a request for an advertisement, wherein the request includes a keyword; and

responsive to the request, delivering the first advertisement in response to the keyword of the request matching a keyword of the one or more keywords associated with the first advertisement.

2. The method of claim 1, wherein associating one or more of the keywords from the reference advertisement with the first advertisement comprises:

determining that the similarity measure between the first advertisement and the reference advertisement exceeds a similarity threshold.

3. The method of claim 2, wherein associating one or more of the keywords from the reference advertisement with the first advertisement comprises:

for each of the one or more keywords, determining a performance of the keyword with the reference advertisement;

comparing the performances of the one or more keywords; and

selecting a keyword from the one or more keywords to associate with the first advertisement based on the performances.

4. The method of claim 3, wherein the selected keyword is a keyword associated with the reference advertisement having a highest performance.

5. The method of claim 3, wherein the performance is a click-through rate.

6. The method of claim 1, wherein the reference advertisement is associated with a subject matter, the method further comprising:

associating the subject matter from the reference advertisement with the first advertisement.

7. The method of claim 1, wherein comparing image data from a first advertisement to image data from a reference advertisement comprises:

comparing patterns in the optical content depicted by the first advertisement to patterns in the optical content from the reference advertisement.

8. The method of claim 1;

wherein determining a similarity measure between the first advertisement and the reference advertisement comprises;

identifying a link included in the first advertisement and a link included in the reference advertisement; and

determining the similarity measure based, at least in part, on whether the link included in the first advertisement matches the link included in the reference advertisement.

9. The method of claim 1, wherein determining a similarity measure between the first advertisement and the reference advertisement comprises:

identifying a target demographic for the first advertisement and a target demographic for the reference advertisement; and

determining the similarity measure based, at least in part, on whether the target demographic for the first advertisement matches the target demographic for the reference advertisement.

10. The method of claim 1, wherein determining a similarity measure between the first advertisement and the reference advertisement comprises:

identifying a document source for the first advertisement and a document source for the reference advertisement; and

determining the similarity measure based, at least in part, on whether the document source for the first advertisement matches the document source for the reference advertisement.

11. The method of claim 1, wherein determining a similarity measure between the first advertisement and the reference advertisement comprises:

identifying a content rating for the first advertisement and a content rating for the reference advertisement; and

determining the similarity measure based, at least in part, on whether the content rating for the first advertisement matches the content rating for the reference advertisement.

12. A computer-implemented method comprising:

comparing image data that specify optical content depicted by a first advertisement to image data that specify optical content depicted by a reference advertisement, wherein the reference advertisement is associated with a concept;

determining, by one or more processors, a similarity measure between the first advertisement and the reference advertisement based on the comparison of image data;

associating the concept from the reference advertisement with the first advertisement based on the similarity measure between the first advertisement and the reference advertisement;

receiving a request for an advertisement, wherein the request is associated with the concept from the reference advertisement; and

responsive to the request, delivering the first advertisement in response to the request being associated with the concept from the reference advertisement.

13. A computer-implemented method comprising:

identifying a first advertisement;

identifying reference advertisements, wherein each reference advertisement is associated with at least one keyword;

comparing image data that specify optical content depicted by the first advertisement with image data that specify optical content depicted by each reference advertisement;

determining, by one or more processors, a set of one or more of the reference advertisements having optical content that matches the optical content of the first advertisement based on the comparison;

determining a predetermined number of highest scoring reference advertisements from the set to associate with the first advertisement;

associating keywords from the predetermined number of highest scoring reference advertisements with the first advertisement; and

using the keywords from the predetermined number of highest scoring reference advertisements to deliver the first advertisement in response to a received advertisement request that includes at least one of the keywords.

14. The method of claim 13, wherein comparing image data that specify optical content depicted by the first advertisement with image data specifying optical content depicted by each reference advertisement comprises:

comparing textual content of the first advertisement with textual content of each reference advertisement.

15. The method of claim 13, wherein comparing image data that specify optical content depicted by the first advertisement with image data that specify optical content depicted by each reference advertisement comprises:

comparing subject matter of the first advertisement with subject matter of each reference advertisement.

16. The method of claim 13, wherein comparing image data that specify optical content depicted by the first advertisement with image data that specify optical content depicted by each reference advertisement comprises:

comparing subject matter of documents to which the first advertisement links with subject matter of documents to which the reference advertisements link.

17. A system, comprising:

one or more processors; and

a computer-readable medium coupled to the one or more processors having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising;

comparing image data that specify optical content depicted by a first advertisement to image data that specify optical content depicted by a reference advertisement, wherein the reference advertisement is associated with at least one keyword;

determining, by one or more processors, a similarity measure between the first advertisement and the reference advertisement based on the comparison of the image data;

associating one or more of the keywords from the reference advertisement with the first advertisement based on the similarity measure between the first advertisement and the reference advertisement;

receiving a request for an advertisement, wherein the request includes a keyword; and

responsive to the request, delivering the first advertisement in response the keyword of the request matching a keyword of the one or more keywords associated with the first advertisement.

18. The system of claim 17, wherein associating one or more of the keywords from the reference advertisement with the first advertisement comprises:

determining that the similarity measure between the first advertisement and the reference advertisement exceeds a similarity threshold.

19. The system of claim 18, wherein associating one or more of the keywords from the reference advertisement with the first advertisement comprises:

for each of the one or more keywords, determining a performance of the keyword with the reference advertisement;

comparing the performances of the one or more keywords; and

selecting a keyword from the one or more keywords to associate with the first advertisement based on the performances.

20. The system of claim 19, wherein the selected keyword is a keyword associated with the reference advertisement having a highest performance.

21. The system of claim 17, wherein determining a similarity measure between the first advertisement and the reference advertisement comprises:

identifying a link included in the first advertisement and a link included in the reference advertisement; and

determining the similarity measure based, at least in part, on whether the link included in the first advertisement matches the link included in the reference advertisement.

22. A system, comprising:

one or more processors; and

a computer-readable medium coupled to the one or more processors having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising;

identifying a first advertisement;

identifying reference advertisement, wherein each reference advertisement is associated with at least one keyword;

comparing image data that specify optical content of the first advertisement with image data that specify optical content of each reference advertisement;

determining, by one or more processors, a set of one or more of the reference advertisements having optical content that matches the optical content of the first advertisement based on the comparison;

determining a predetermined number of highest scoring reference advertisements from the set to associate with the first advertisement;

associating keywords from the predetermined number of highest scoring reference advertisements with the first advertisement; and

using the keywords from the predetermined number of highest scoring reference advertisements to deliver the first advertisement in response to a received advertisement request that includes at least one of the keywords.

23. A computer storage medium encoded with a computer program, the program comprising instructions that when executed by data processing apparatus cause the data processing apparatus to perform operations, comprising:

comparing image data that specify optical content from a first advertisement to image data that specify optical content from a reference advertisement, wherein the reference advertisement is associated with at least one keyword;

determining, by one or more processors, a similarity measure between the first advertisement and the reference advertisement based on the comparison of the image data;

associating one or more of the keywords from the reference advertisement with the first advertisement based on the similarity measure between the first advertisement and the reference advertisement;

receiving a request for an advertisement, wherein the request includes a keyword; and

responsive to the request, delivering the first advertisement in response to the keyword of the request matching a keyword of the one or more keywords associated with the first advertisement.