Aberrant Behavior Detection in Time Series for Network Monitoring
LISA '00 Proceedings of the 14th USENIX conference on System administration
Why we twitter: understanding microblogging usage and communities
Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis
Proceedings of the first workshop on Online social networks
Proceedings of the 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Twitter power: Tweets as electronic word of mouth
Journal of the American Society for Information Science and Technology
Earthquake shakes Twitter users: real-time event detection by social sensors
Proceedings of the 19th international conference on World wide web
Listen to me if you can: tracking user experience of mobile network on social media
IMC '10 Proceedings of the 10th ACM SIGCOMM conference on Internet measurement
Tweetflows: flexible workflows with twitter
Proceedings of the 3rd International Workshop on Principles of Engineering Service-Oriented Systems
Outage detection via real-time social stream analysis: leveraging the power of online complaints
Proceedings of the 21st international conference companion on World Wide Web
Scaling microblogging services with divergent traffic demands
Middleware'11 Proceedings of the 12th ACM/IFIP/USENIX international conference on Middleware
Scaling microblogging services with divergent traffic demands
Proceedings of the 12th International Middleware Conference
Hi-index | 0.00 |
Real-time micro-blogging services such as Twitter are widely recognized for their social dynamics--how they both encapsulate a social graph and propagate information across it. However, the content of this information is equally interesting since it frequently reflects individual experiences with a broad variety of real-time events. Indeed, events of broad interest are commonly revealed in correlated spikes of semantically-related posting activity. In this paper, we explore one such application this of phenomenon: using Twitter data to infer on-line Internet service availability. We show that simple techniques are sufficient to extract key semantic content from "tweets" (i.e., service X is down) and also filter out extraneous noise. We demonstrate the efficacy of this approach at identifying a range of large-scale service outages in 2009 for popular services such as Gmail, Bing and PayPal.