log in  |  register  |  feedback?  |  help  |  web accessibility
Logo
Discovering Key Moments in Social Media Streams
Wednesday, February 24, 2016, 11:00 am-12:00 pm Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)
Abstract
We introduce a general technique, called LABurst, for identifying key moments, or moments of high impact, in social media streams without the need for domain-specific information or seed keywords.  We leverage machine learning to model temporal patterns around bursts in Twitter's unfiltered public sample stream and build a classifier to identify tokens experiencing these bursts.  LABurst performs competitively with existing burst detection techniques while simultaneously providing insight into and detection of unanticipated moments.  To demonstrate our approach's potential, we compare two baseline event-detection algorithms with our language-agnostic algorithm to detect key moments across three major sporting competitions: 2013 World Series, 2014 Super Bowl, and 2014 World Cup.  Our results show LABurst outperforms a time series analysis baseline and is competitive with a domain-specific baseline even though we operate without any domain knowledge.  We then go further by transferring LABurst's models learned in the sports domain to the task of identifying earthquakes in Japan and show our method detects large spikes in earthquake-related tokens within two minutes of the actual event.
Bio

Cody Buntain is a PhD student in CS at the University of Maryland. His primary research is into social networks with specific interests in how social media can describe events on the ground. His dissertation work focuses on developing algorithms to identify this event information and assess credibility in real time.

This talk is organized by Naomi Feldman