PROVIDING CONTEXT FOR WEB ARTICLES

• US 20090319449A1
• Filed: 06/21/2008
• Published: 12/24/2009
• Est. Priority Date: 06/21/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Konig, Arnd C., Belenko, Dmitriy A., Hurst, Matthew F., Basu, Sumit, Gamon, Michael, Fisher, Danyel A.
• Granted Patent: US 8,630,972 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 16/94 Hypermedia Hyperlinking G06F40/134

Abstract

An overwhelming number of articles are available everyday via the internet. Unfortunately, it is impossible to peruse more than a handful, and it is difficult to ascertain an article'"'"'s social context. The techniques disclosed herein address this problem by harnessing implicit and explicit contextual information from social media. By extracting text surrounding a hyperlink to an article in a post and assessing the article as a function of content surrounding the hyperlink, an article'"'"'s social context is determined and presented. Additionally, articles that are sufficiently similar in content may be grouped to establish a many-to-one relationship between posts and an article, creating a more accurate assessment.

Claims

1. A method for making an assessment of an article comprising:

assessing the article as a function of meaning of content surrounding one or more hyperlinks pointing to the article in view of contemporary lexicon.

2. The method of claim 1, comprising extracting the hyperlink from a post.

3. The method of claim 2, comprising acquiring the post from syndicated feed crawlers that crawl social media.

4. The method of claim 1, comprising retrieving a full text page comprising the article.

5. The method of claim 4, comprising using an automatic method to assess the content surrounding a hyperlink pointing to an article.

6. The method of claim 1, comprising modifying the hyperlink pointing to the article when a different hyperlink points to a different article comprising sufficiently similar content.

7. The method of claim 6, comprising detecting articles that are sufficiently similar in content wherein

$\frac{S}{}$

( a i , a j ) size 

( a i )

exceeds a predetermined threshold, where S(a_i,a_j) is the number of terms of the largest set of infrequent terms between articles a_i and a_j and size (a_i) is the number of terms in the larger of articles a_i and a_j.

8. The method of claim 6, comprising detecting articles that are sufficiently similar in content wherein the fraction of words in W(a_i) that are covered by at least one set of term sequences that occur in both articles a_i and a_j exceeds a predetermined threshold, where W(a_i) is the substring of all terms in the article a_i not covered by frequent terms

9. The method of claim 1, comprising:

presenting the article to a user, and

presenting the assessment of the article to the user.

10. The method of claim 1, the assessing the article comprising:

making an assessment of the article as a function of some predetermined criteria.

11. The method of claim 10, the predetermined criteria comprising an emotional charge.

12. The method of claim 10, the predetermined criteria comprising a number of hyperlinks pointing to the article.

13. The method of claim 10, comprising training an annotator to make the assessment.

14. A method for making an assessment of an article comprising:

monitoring social media and acquiring a post;

retrieving a first article corresponding to a first hyperlink in the post pointing to the first article;

updating the first hyperlink pointing to the first article when a second hyperlink points to a second article comprising sufficiently similar content;

upon determining a modified first article, assessing the modified first article as a function of meaning of content surrounding the first hyperlink in view of contemporary lexicon; and

presenting the assessment of the modified first article, the modified first article, and the post comprising the first hyperlink that pointed to the first article.

15. The method of claim 14, comprising assessing the modified first article as a function of content surrounding the first hyperlink and second hyperlink.

16. The method of claim 14, comprising presenting the assessment of the article, the article, and the post comprising the first hyperlink pointing to a first article and second hyperlink pointing to a second article.

17. The method of claim 14, comprising assessing the article for emotional charge.

18. The method of claim 17, comprising using a maximum entropy classifier to make the assessment.

19. A system for assessing an article, the system comprising:

an acquisition component configured to acquire a post;

an article scraper component configured to retrieve articles;

a de-duplication component configured to detect articles comprising sufficiently similar content;

an assessment component configured to create an assessment of the article as a function of meaning of content surrounding a hyperlink pointing to the article in the post in view of contemporary lexicon;

a servicing component configured to allow the post, the assessment, and the article to be retrieved; and

a user interface configured to present the article and present an assessment of the article.

20. The system of claim 19, the assessment a function of some predetermined criteria.