Online Processing of Social Media Data for Emergency Management

Social media offers an opportunity for emergency management to identify issues that need immediate reaction. To support the effective use of social media, an analysis approach is needed to identify crisis-related hotspots. We consider in this investigation the analysis of social media (i.e., Twitter, Flickr and YouTube) to support emergency management by identifying sub-events. Sub-events are significant hotspots that are of importance for emergency management tasks. Aiming at sub-event detection, recognition and tracking, the data is processed online in real-time. We introduce an incremental feature selection mechanism to identify meaningful terms and use an online clustering algorithm to uncover sub-events on-the-fly. Initial experiments are based on tweets enriched with Flickr and YouTube data collected during Hurricane Sandy. They show the potential of the proposed approach to monitor sub-events for real-world emergency situations.

[1]  Hermann Hellwagner,et al.  Automatic Identification of Crisis-Related Sub-events Using Clustering , 2012, 2012 11th International Conference on Machine Learning and Applications.

[2]  Tao Li,et al.  A Participant-based Approach for Event Summarization Using Twitter Streams , 2013, NAACL.

[3]  Christopher D. Manning,et al.  Introduction to Information Retrieval , 2010, J. Assoc. Inf. Sci. Technol..

[4]  M. Osborne,et al.  Bieber no more : First Story Detection using Twitter and Wikipedia , 2012 .

[5]  Huan Liu,et al.  Toward integrating feature selection algorithms for classification and clustering , 2005, IEEE Transactions on Knowledge and Data Engineering.

[6]  Abdelhamid Bouchachia,et al.  Incremental Learning Based on Growing Gaussian Mixture Models , 2011, 2011 10th International Conference on Machine Learning and Applications and Workshops.

[7]  M. F. Porter,et al.  An algorithm for suffix stripping , 1997 .

[8]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[9]  Deepayan Chakrabarti,et al.  Event Summarization Using Tweets , 2011, ICWSM.

[10]  Thorsten Brants,et al.  A System for new event detection , 2003, SIGIR.

[11]  Leysia Palen,et al.  Microblogging during two natural hazards events: what twitter may contribute to situational awareness , 2010, CHI.

[12]  Geert-Jan Houben,et al.  Twitcident: fighting fire with information from social web streams , 2012, WWW.

[13]  Martin Ester,et al.  Frequent term-based text clustering , 2002, KDD.

[14]  P. Rousseeuw.  Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[15]  Grigorios Tsoumakas,et al.  On the Utility of Incremental Feature Selection for the Classification of Textual Data Streams , 2005, Panhellenic Conference on Informatics.

[16]  Philip D. Cha,et al.  Fundamentals of Signals and Systems: A Building Block Approach , 2005 .

[17]  Wei-Ying Ma,et al.  An Evaluation on Feature Selection for Text Clustering , 2003, ICML.

[18]  Ramesh Nallapati,et al.  Event threading within news topics , 2004, CIKM '04.

[19]  Christiane Fellbaum,et al.  Book Reviews: WordNet: An Electronic Lexical Database , 1999, CL.

[20]  Vassilis Kostakos,et al.  A real-time social media aggregation tool: Reflections from five large-scale events , 2011 .

[21]  Michael S. Bernstein,et al.  Twitinfo: aggregating and visualizing microblogs for event exploration , 2011, CHI.

[22]  Roland Mittermeir,et al.  Towards Incremental Fuzzy Classifiers , 2006, Soft Comput..

[23]  Hao Wang,et al.  Online Streaming Feature Selection , 2010, ICML.