RELATED NEWS ARTICLES

• US 20130132401A1
• Filed: 11/17/2011
• Published: 05/23/2013
• Est. Priority Date: 11/17/2011
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Zheng, Zhaohui, Chang, Yi, Kolari, Pranam, Moon, Taesup, Wang, Xuanhui, Lv, Yuanhua
• Granted Patent: US 8,713,028 B2
• Time in Patent Office: Days
• US Class Current: 707/748
• CPC Class Codes: G06F 16/9032 Query formulation
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Methods, systems, and computer programs are presented for providing internet content, such as related news articles. One method includes an operation for defining a plurality of candidates based on a seed. For each candidate, scores are calculated for relevance, novelty, connection clarity, and transition smoothness. The score for connection clarity is based on a relevance score of the intersection between the words in the seed and the words in each of the candidates. Further, the score for transition smoothness measures the interest in reading each candidate when transitioning from the seed to the candidate. For each candidate, a relatedness score is calculated based on the calculated scores for relevance, novelty, connection clarity, and transition smoothness. In addition, at least one of the candidates is selected based on their relatedness scores for presentation to the user.

Claims

1. A method for providing internet content, the method comprising:

defining a plurality of candidates based on a seed;

for each candidate, calculating scores for relevance, novelty, connection clarity, and transition smoothness, wherein the score for connection clarity is based on a relevance score of an intersection between the seed and each candidate, wherein the score for transition smoothness measures an interest in reading each candidate when transitioning from the seed to each candidate;

for each candidate, calculating a relatedness score based on the calculated scores for relevance, novelty, connection clarity, and transition smoothness; and

selecting at least one candidate from the plurality of candidates based on the relatedness scores, wherein at least one operation of the method is executed through a processor.

2. The method as recited in claim 1, wherein the score for connection clarity is further based on topic distributions of the seed and each candidate.

3. The method as recited in claim 1, wherein the score for connection clarity is calculated with a formula of

4. The method as recited in claim 1, wherein the score for connection clarity is calculated with a formula of

5. The method as recited in claim 1, wherein the score for transition smoothness is based on comparing contexts for (s−

d) and (d−

s).

6. The method as recited in claim 1, wherein the score for transition smoothness is calculated with a formula of

7. The method as recited in claim 1, wherein the score for transition smoothness is calculated with a formula of

8. The method as recited in claim 1 further including:

eliminating candidates that are substantially identical to the seed.

9. The method as recited in claim 1, wherein calculating the relatedness score utilizes a learning to rank framework that combines relevance, novelty, connection clarity, and transition smoothness.

10. The method as recited in claim 1, further including:

presenting the selected at least one candidate from the plurality of candidates to a user.

11. A computer program embedded in a non-transitory computer-readable storage medium, when executed by one or more processors, for providing internet content, the computer program comprising:

program instructions for defining a plurality of candidates based on a seed;

program instructions for calculating, for each candidate, scores for relevance, novelty, connection clarity, and transition smoothness, wherein the score for connection clarity is based on a relevance score of an intersection between the seed and each candidate, wherein the score for transition smoothness measures an interest in reading each candidate when transitioning from the seed to each candidate;

program instructions for calculating, for each candidate, a relatedness score based on the calculated scores for relevance, novelty, connection clarity, and transition smoothness; and

program instructions for selecting at least one candidate from the plurality of candidates based on the relatedness scores.

12. The computer program as recited in claim 11, wherein the program instructions for defining the plurality of candidates further includes:

program instructions for using a standard information retrieval model to retrieve a set of documents using the seed as a query; and

program instructions for removing redundant articles to obtain the plurality of candidates.

13. The computer program as recited in claim 11, wherein the relatedness score is calculating using a retrieval model selected from a group consisting of a vector space model, a classic probabilistic model, and a language model.

14. The computer program as recited in claim 11, wherein the seed includes title, abstract, and body.

15. The computer program as recited in claim 11, wherein the score for connection clarity is further based on topic distributions of the seed and each candidate.

16. A method for providing news articles, the method comprising:

defining a plurality of news candidates based on a seed article selected by a user;

for each news candidate, calculating scores for relevance, novelty, connection clarity, and transition smoothness, wherein the score for connection clarity is based on a relevance score of an intersection between the seed article and each news candidate, wherein the score for transition smoothness measures an interest in reading each news candidate when transitioning from the seed article to each news candidate;

for each news candidate, calculating a relatedness score based on the calculated scores for relevance, novelty, connection clarity, and transition smoothness; and

selecting based on the relatedness scores a subset of news candidates from the plurality of news candidates for presentation to the user, wherein at least one operation of the method is executed through a processor.

17. The method as recited in claim 16, wherein the score for connection clarity is calculated with a formula of

18. The method as recited in claim 16, wherein the score for connection clarity score is calculated with a formula of

19. The method as recited in claim 16, wherein the score for transition smoothness is calculated with a formula of

20. The method as recited in claim 16, wherein operations of the method are performed by a computer program when executed by one or more processors, the computer program being embedded in a non-transitory computer-readable storage medium.