Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/www2010/paper/main/788> ?p ?o. }
Showing items 1 to 11 of
11
with 100 items per page.
- 788 creator kristina-lerman.
- 788 creator tad-hogg.
- 788 type InProceedings.
- 788 label "Using a Model of Social Dynamics to Predict Popularity of News".
- 788 sameAs 788.
- 788 abstract "Popularity in social media is unequally distributed with few of the items, such as blog posts, stories on the social news portal Digg, images on the photo-sharing site Flickr, etc., grabbing a disproportionate number of votes or views. Predicting which items will become popular is critically important for both social media companies that host user-generated content and consumers of that content. Accurate and timely prediction would enable the companies to maximize revenue through differential pricing for access to content or ad placement. Prediction would give consumers an important tool for dealing with the ever-growing amount of content. Predicting popularity of content in social media, however, is a challenging problem. In a landmark study, Salganik et al.~\cite{Salganik06} showed that social influence, or the ability to see the choices of others, causes popularity to be both unequally distributed and unpredictable. We argue that observability of user behavior on many social media sites allows us to construct a model of social dynamics of users on the site. The model can in turn be used to predict long-term popularity of content based on how its early popularity changes in time. We validate this claim on the social news portal Digg. We use the model of social voting developed in earlier works to predict how many votes stories will receive on Digg based on the users' early reaction to them.".
- 788 hasAuthorList authorList.
- 788 isPartOf proceedings.
- 788 keyword "Social data analysis".
- 788 keyword "analytics".
- 788 title "Using a Model of Social Dynamics to Predict Popularity of News".