Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/www2012/poster/77> ?p ?o. }
Showing items 1 to 15 of
15
with 100 items per page.
- 77 creator belle-tseng.
- 77 creator jiang-bian.
- 77 creator xuanhui-wang.
- 77 creator yi-chang.
- 77 type InProceedings.
- 77 label "Model News Relatedness through User Comments".
- 77 sameAs 77.
- 77 abstract "Recommending related news articles aims to engage with users afterthey read current ones and can improve user retention for newsproviders. Most of previous works on news relatedness focus on thearticle texts. In this paper, we study the benefit ofuser-generated comments on modeling news relatedness. Commentscontain rich text information which is provided by commenters andrated by readers with thumb-up or thumb-down, but the quality ofindividual comments varies widely. We compare different ways ofcapturing relatedness by leveraging both text and user interactioninformation in comments. Our evaluation based on an editorial dataset demonstrates that the text information in comments is veryeffective to model relatedness while community rating is quitepredictive of the comment quality.".
- 77 hasAuthorList authorList.
- 77 isPartOf proceedings.
- 77 isPartOf proceedings.
- 77 keyword "Community rating".
- 77 keyword "News Relatedness".
- 77 keyword "User comments".
- 77 title "Model News Relatedness through User Comments".