Tow Center Presents Carlos Castillo | Social Media News Mining
Thursday, November 14, 2013 from 6:30 PM to 8:00 PM (EST)
As part of its Speaker Series on Computational Journalism, the Tow Center presents Carlos Castillo.
The evening includes a 45-minute lecture by Carlos Castillo followed by an interview with Gilad Lotan and a public Q&A session. Refreshments will be served.
Social Media News Mining and Automatic Content Analysis of News
Abstract: We start by considering social media users who share a specific online news article on a social media platform. This group is the “news crowd” of the article, an analogy with the crowd of passers-by that gathers around any incident on a busy street. First, we study both the volume and the diversity of the social media conversations of this crowd around the news item, to produce improved predictions of its impact. Second, we follow the crowd during several days in order to find out other news which may be related to the original one.
The next step is to look at the content of news. We analyze news from 140 television channels in the US by extracting the text of their closed captions, and from 60+ international news websites by extracting the title and content of news articles. We perform automatic content analysis to quantify differences in their treatment of news, and to measure gatekeeping, coverage and statement bias. We close the loop by looking at differences on the treatment of news subjects by social media and traditional news media.
This talk is based on joint work with Nasir Khan and Mohammed El-Haddad (Al Jazeera), Gianmarco de Francisci Morales, Mounia Lalmas, Janette Lehmann, Marcelo Mendoza and Diego Sáez-Trumper (Yahoo!), Jürgen Pfeffer (CMU),Matt Stempeck and Ethan Zuckerman (MIT Center for Civic Media), and Sandra González-Bailón (University of Pennsylvania).
Carlos Castillo is a Senior Scientist at the Qatar Computing Research Institute in Doha. Before moving to Qatar, he worked at Yahoo! Research for six years. He is a web miner with a background on information retrieval, and has been influential in the areas of adversarial web search and web content quality and credibility. He is an active researcher with more than 40+ publications in top-tier international conferences and journals and 4100+ citations. His current research focuses in the application of web mining methods to problems in the domain of on-line news (in collaboration with Al Jazeera) and humanitarian crises. More about “ChaTo” on http://www.chato.cl/research/social_media_news/
Gilad Lotan is the Chief Data Scientist at betaworks, a technology company that operates as a studio, building new products, growing companies and seed investing. Previously, Gilad ran the data team at SocialFlow and built visualizations at Microsoft's FUSE Labs. His past work includes ‘Retweet Revolution’, visualizing the flow of information during the 2009 #IranElection riots, and a study investigating the relationship between mainstream media and social media channels during the Tunisian and Egyptian revolutions. Gilad is an advisor for media entities and startups. His work has been covered by the New York Times, the Guardian, Fast Company and the Atlantic Wire and published across a wide range of academic journals.