Dynamic captions from social streams

• US 9,317,583 B2
• Filed: 10/05/2012
• Issued: 04/19/2016
• Est. Priority Date: 10/05/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Alonso, Omar, Kalyanaraman, Shankar
• Primary Examiner(s): MORRISON, JAY A
• Application Number:
  US13/645,533
• Publication Number:
  US 20140101145A1
• Time in Patent Office: 1,292 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing

Abstract

Architecture that augments a search result entry caption with keywords related to topics currently being shared and discussed in other social network information sources. This can provide a much better idea of the content of the website or webpage. The architecture obtains a link from updates of social topics of social network information sources (e.g., social networks for people places professionals, etc.), extracts title content of a document title associated with the link, compares the title content to document text for similarity to create non-duplicative content, creates keywords related to popular social topics from the non-duplicative content, and then augments a search result entry of a search result page with the keywords of the popular social topics.

Claims

1. A system, comprising:

a search result entry, presented on a search result page from a user search performed using a search engine;

an augmentation of the search result entry with social content fragments presented with the search result entry on the search result page, the augmentation obtained via links to social network information sources, the social content fragments are automatically updated on the search result page based on updated social content of the social network information sources, the social content fragments are keywords obtained from a comparison of title content of the links to document text to determine non-duplicative content; and

a hardware microprocessor that executes computer-executable instructions associated with at least the augmentation.

2. The system of claim 1, wherein the links are to documents from which document titles are extracted and processed to create the non-duplicative content.

3. The system of claim 2, wherein the document titles are compared to link text using a similarity measure to generate similarity scores.

4. The system of claim 2, wherein the non-duplicative content is processed to create the social content fragments presented in association with the search result entry.

5. The system of claim 1, wherein the social network information sources include at least one of a social network for places, a social network for professionals, a social network for people, a question and answer source, or a microblogging service.

6. The system of claim 1, wherein the social content fragments are derived from at least one of tips, conversations, or questions of the social network information sources.

7. The system of claim 1, wherein the social content fragments are updated dynamically based on the updated social content.

8. The system of claim 1, wherein the social content fragments are presented in association with a caption of the search result entry.

9. A method performed by a computer system executing machine-readable instructions, the method comprising acts of:

performing a search using a search engine to obtain a search result entry that is presented on a search result page;

obtaining a link from updates of social data of social network information sources;

extracting title content of a document title associated with the link;

comparing the title content to text associated with the link to create keywords related to popular topics; and

augmenting information of the search result entry on the search result page with the keywords of the popular topics corresponding to the updates of the social data of the social network information sources.

10. The method of claim 9, further comprising comparing the title content to text associated with the link to create non-duplicative content.

11. The method of claim 10, further comprising creating keywords related to popular topics from the non-duplicative content.

12. The method of claim 9, further comprising comparing the title content to the text using a similarity measure to create a similarity score.

13. The method of claim 9, further comprising augmenting the search result entry of the search result page with the keywords.

14. The method of claim 9, further comprising augmenting a caption fly-out of the search result entry of the search result page with the keywords.

15. The method of claim 9, further comprising augmenting a specific vertical of the search result page with the keywords.

16. A method performed by a computer system executing machine-readable instructions, the method comprising acts of:

performing a user search using a search engine to obtain a search result entry presented on a search result page;

obtaining a link from updates of social topics of social network information sources;

extracting title content of a document title associated with the link;

comparing the title content to document text for similarity to create non-duplicative content;

creating keywords related to popular social topics from the non-duplicative content; and

augmenting the search result entry of the search result page with the keywords of the popular social topics corresponding to the updates of the social topics of the social network information sources.

17. The method of claim 16, further comprising augmenting a caption of the search result entry with the keywords as part of caption generation.

18. The method of claim 16, further comprising analyzing datasets of the social network information sources of a given time span to compute the popular social topics.

19. The method of claim 16, further comprising aging out a keyword over time based on aging criteria.

20. The method of claim 16, further comprising obtaining the link to updates of social network information sources that include at least one of a social network for places, a social network for professionals, a social network for people, a question and answer source, or a microblogging service.