Enhancing sitelinks with creative content

• US 10,650,066 B2
• Filed: 03/15/2013
• Issued: 05/12/2020
• Est. Priority Date: 01/31/2013
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Vaish, Vaibhav, Ramachandran, Venky, Sisson, David Philip, Kandhan, Ramakrishnan, Adiddam, Pramod, Marur, Vinod, Garg, Gaurav
• Primary Examiner(s): Ubale, Gautam
• Application Number:
  US13/840,380
• Publication Number:
  US 20140214790A1
• Time in Patent Office: 2,615 Days
• Field of Search: 705 14
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Methods and systems for enhancing online content with creative text relevant to the online content are provided. A plurality of candidate sitelinks is identified in response to a user search for online content. Each sitelink has associated with it a plurality of candidate creatives with which the sitelink may be presented to the user. The creatives are canonicalized to form clusters of candidate creatives. The sitelinks are also canonicalized. The creatives are matched to the candidate canonicalized sitelinks so as to provide enhanced sitelinks having increased relevance to the user search.

Claims

1. A computer-implemented method of matching sitelinks with content items, comprising:

identifying, by a data processing system having one or more processors and memory, a primary result associated with a content provider, responsive to a search query from a client device;

identifying, by the data processing system, for presentation with the primary result, a sitelink associated with the content provider for the primary result and including a sitelink uniform resource locator (URL) indexed in a sitelink database, the sitelink URL including a sitelink parameter;

generating, by the data processing system, a canonicalized sitelink URL for the sitelink URL by removing the sitelink parameter;

identifying, by the data processing system, for presentation with the primary result, from a content item database for the content provider, a plurality of content items associated with;

a first URL including a first parameter, a second URL including a second parameter, and a third URL including a third parameter;

generating, by the data processing system, a first canonicalized URL for the first URL by removing the first content item parameter;

generating, by the data processing system, a second canonicalized URL for the second URL by removing the second content item parameter;

generating, by the data processing system, a third canonicalized URL for the third URL by removing the third content item;

determining, by the data processing system, that the first canonicalized URL is the same as the second canonicalized URL but differs from the third canonicalized URL;

identifying, by the data processing system, both the first canonicalized URL and the second canonicalized URL as associated with a first content item from the plurality of content items based on determining that the first canonicalized URL is the same as the second canonicalized URL;

identifying, by the data processing system, the third canonicalized URL as associated with a second content item different from the first content item from the plurality of content items based on determining that the first canonicalized URL differs from the third canonicalized URL;

generating, by the data processing system, a first content item URL group for the first content item including the first canonicalized URL and the second canonicalized URL and a second content item URL group for the second content item including the third canonicalized URL, responsive to identifying both the first canonicalized content item URL and the second canonicalized URL as associated with the first content item and identifying the third canonicalized URL as associated with the second content item;

determining, by the data processing system, a first score indicating a similarity between text of the first content item for the first content item URL group and text of the sitelink and a second score indicating a similarity between text of the second content item for the second content item URL group and the text of the sitelink;

matching, by the data processing system, the sitelink with the first content item based on a comparison between the first score and the second score; and

generating, by the data processing system, for presentation on the client device responsive to the search query, a results interface page associated with the content provider including the primary result and one or more secondary results including the sitelink, and the text of the first content item.

2. The method of claim 1, further comprising pruning a candidate set of content items matched with sitelinks by removing at least one of duplicate content items and redundant content items.

3. The method of claim 1, wherein the filter rules include at least one of language rules, geographic rules, user device rules, platform rules, and campaign rules.

4. The method of claim 1, comprising:

indexing, by the data processing system, a plurality of sitelink URLs with and without a URL parameter associated with each of the plurality of sitelink URLs;

comparing, by the data processing system, landing pages of each of the plurality of sitelink URLs; and

removing, by the data processing system, the URL parameter from each of the plurality of sitelink URLs responsive to determining that the landing pages of each of the plurality of sitelink URLs match.

5. The method of claim 1, wherein generating the canonicalized sitelink URL includes processing webmaster supplied rules specifying the relevance of the sitelink parameter.

6. The method of claim 1, wherein generating the canonicalized URL includes comparing contents of a first landing page associated with the first canonicalized URL to identify similarities with a second landing page associated with the second canonicalized URL.

7. The method of claim 1, wherein generating the first canonicalized URL includes processing webmaster supplied rules indicating the relevance of content item parameter.

8. The method of claim 1, wherein selecting the content item further comprises matching a plurality of sitelinks with an as-yet unmatched content item having a highest estimated score from amongst unmatched content items.

9. The method of claim 1, wherein selecting the content item further comprises determining maximize total estimated scores amongst a plurality of sitelinks and as-yet unmatched content item.

10. The method of claim 1, comprising determining, by the data processing system, an impression score indicating a number of impressions for the content item.

11. The method of claim 1, comprising determining, by the data processing system, an inverse-document-frequency (IDF) score indicating similarity of terms between the text of the sitelink associated with the sitelink URL and the text of the first content item associated with the first URL.

12. A computer system to match sitelinks with content items, comprising:

a processor; and

a computer-readable storage device having encoded thereon computer readable instructions that are executable by the processor to perform functions comprising;

identifying a primary result associated with a content provider, responsive to a search query from a client device;

identifying, for presentation with the primary result, a sitelink associated with the content provider for the primary result and including a sitelink uniform resource locator (URL) indexed in a sitelink database, the sitelink URL including a sitelink parameter;

generating a canonicalized sitelink URL for the sitelink URL by removing the sitelink parameter;

identifying, for presentation with the primary result, from a content item database for the content provider, a plurality of content items associated with;

a first URL including a first parameter, a second URL including a second parameter, and a third URL including a third parameter;

generating a first canonicalized URL for the first URL by removing the first content item parameter;

generating a second canonicalized URL for the second URL by removing the second content item parameter;

generating a third canonicalized URL for the third URL by removing the third content item;

determining that the first canonicalized URL is the same as the second canonicalized URL but differs from the third canonicalized URL;

identifying both the first canonicalized URL and the second canonicalized URL as associated with a first content item from the plurality of content items based on determining that the first canonicalized URL is the same as the second canonicalized URL;

identifying the third canonicalized URL as associated with a second content item different from the first content item from the plurality of content items based on determining that the first canonicalized URL differs from the third canonicalized URL;

generating a first content item URL group for the first content item including the first canonicalized URL and the second canonicalized URL and a second content item URL group for the second content item including the third canonicalized URL, responsive to identifying both the first canonicalized content item URL and the second canonicalized URL as associated with the first content item and identifying the third canonicalized URL as associated with the second content item;

determining a first score indicating a similarity between text of the first content item for the first content item URL group and text of the sitelink and a second score indicating a similarity between text of the second content item for the second content item URL group and the text of the sitelink;

matching the sitelink with the first content item based on

a comparison between the first score and the second score; and

generating, for presentation on the client device responsive to the search query, a results interface page associated with the content provider including the primary result and one or more secondary results including the sitelink, and the text of the first content item.

13. The computer system of claim 12, wherein the computer-executable instructions cause the processor to remove at least one of duplicate content items and redundant content items.

14. The computer system of claim 12, wherein the filter rules include at least one of language rules, geographic rules, user device rules, platform rules, and campaign rules.

15. The computer system of claim 12, wherein the computer-executable instructions cause the processor to index a plurality of sitelink URLs with and without a URL parameter associated with each of the plurality of sitelink URLs, compare landing pages of each of the plurality of sitelink URLs, and remove the URL parameter from each of the plurality of sitelink URLs responsive to the determination that the landing pages of each of the plurality of sitelink URLs match.

16. The computer system of claim 12, wherein the computer-executable instructions cause the processor to generate the canonicalized sitelink URL, through applying webmaster supplied rules indicating the relevance of the sitelink parameter.

17. The computer system of claim 12, wherein computer-executable instructions cause the processor to compare contents of a landing page associated with the first canonicalized URL to identify similarities with a second landing page associated with the second canonicalized URL.

18. The computer system of claim 12, wherein computer-executable instructions cause the processor to select the content item based by matching a plurality of sitelinks with an as-yet unmatched content item having a highest estimated score from amongst unmatched content items.

19. The computer system of claim 12, wherein computer-executable instructions cause the processor to select the content item by determining maximize total estimated scores amongst a plurality of received sitelinks and as-yet unmatched content items.

20. The computer system of claim 12, wherein computer-executable instructions cause the processor to determine an impression score indicating a number of impressions associated with the content item.

21. The computer system of claim 12, wherein computer-executable instructions cause the processor to determine an inverse-document-frequency (IDF) score indicating similarity of terms between the text of the sitelink associated with the sitelink URL and the text of the content item text associated with the first URL.

22. Non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein, when executed by at least one processor associated with a first computing device and a memory device, the computer executable instructions cause the processor to:

identify a primary result associated with a content provider, responsive to a search query from a client device;

identify, for presentation with the primary result, a sitelink associated with the content provider for the primary result and including a sitelink uniform resource locator (URL) indexed in a sitelink database, the sitelink URL including a sitelink parameter;

generate a canonicalized sitelink URL for the sitelink URL by removing the sitelink parameter;

identify, for presentation with the primary result, from a content item database for the content provider, a plurality of content items associated with;

a first URL including a first parameter, a second URL including a second parameter, and a third URL including a third parameter;

generate a first canonicalized URL for the first URL by removing the first content item parameter;

generate a second canonicalized URL for the second URL by removing the second content item parameter;

generate a third canonicalized URL for the third URL by removing the third content item;

determine that the first canonicalized URL is the same as the second canonicalized URL but differs from the third canonicalized URL;

identify both the first canonicalized URL and the second canonicalized URL as associated with a first content item from the plurality of content items based on determining that the first canonicalized URL is the same as the second canonicalized URL;

identify the third canonicalized URL as associated with a second content item different from the first content item from the plurality of content items based on determining that the first canonicalized URL differs from the third canonicalized URL;

generate a first content item URL group for the first content item including the first canonicalized URL and the second canonicalized URL and a second content item URL group for the second content item including the third canonicalized URL, responsive to identifying both the first canonicalized content item URL and the second canonicalized URL as associated with the first content item and identifying the third canonicalized URL as associated with the second content item;

determine a first score indicating a similarity between text of the first content item for the first content item URL group and text of the sitelink and a second score indicating a similarity between text of the second content item for the second content item URL group and the text of the sitelink;

match the sitelink with the first content item based on a comparison between the first score and the second score; and

generate, for presentation on the client device responsive to the search query, a results interface page associated with the content provider including the primary result and one or more secondary results including the sitelink, and the text of the first content item.

23. The non-transitory computer-readable storage media of claim 22, wherein the computer-executable instructions cause the processor to remove at least one of duplicate content items and redundant content items.

24. The non-transitory computer-readable storage media of claim 22, wherein the filter rules include at least one of language rules, geographic rules, user device rules, platform rules, and campaign rules.

25. The non-transitory computer-readable storage media of claim 22, wherein the computer-executable instructions cause the processor to index a plurality of sitelink URLs with and without a URL parameter associated with each of the plurality of sitelink URLs, compare landing pages of each of the plurality of sitelink URLs, and remove the URL parameter from each of the plurality of sitelink URLs responsive to the determination that the landing pages of each of the plurality of sitelink URLs match.

26. The non-transitory computer-readable storage media of claim 22, wherein the computer-executable instructions cause the processor to generate the canonicalized sitelink URL, through applying webmaster supplied rules indicating the relevance of the sitelink parameter.

27. The non-transitory computer-readable storage media of claim 22, wherein computer-executable instructions cause the processor to compare contents of a landing page associated with the first canonicalized URL to identify similarities with a second landing page associated with the second canonicalized URL.

28. The non-transitory computer-readable storage media of claim 22, wherein computer-executable instructions cause the processor to select the content item by matching a plurality of sitelinks with an as-yet unmatched content item having a highest estimated score from amongst unmatched content items.

29. The non-transitory computer-readable storage media of claim 22, wherein computer-executable instructions cause the processor to select the content item by determining maximize total estimated scores amongst a plurality of received sitelinks and as-yet unmatched content items.

30. The non-transitory computer-readable storage media of claim 22, wherein computer-executable instructions cause the processor to determine an impression score indicating a number of impressions associated with the content item.

31. The non-transitory computer-readable storage media of claim 22, wherein computer-executable instructions cause the processor to determine an inverse-document-frequency (IDF) score indicating similarity of terms between the text of the sitelink associated with the sitelink URL and the text of the content item associated with the first URL.