EMERSE@PSU

Enhanced Messaging for Emergency Response
The Pennsylvania State University

Welcome to the EMERSE Project at IST
Latest News: Website has been EMERSE'd!

 

Welcome!

The tragic 7.0 Earthquake in Haiti has mobilized a worldwide relief effort, especially through novel uses of cyberspace.

Even though the earthquake damaged much of the communication infrastructure in Haiti, Haiti's internet connectivity remains robust since most Haitian internet service providers use satellite, rather than damaged undersea fiber optic cable links, to connect to the Internet. Consequently, relief workers, Haitians, and non-governmental organizations (NGO's) have extensively used Tweets to spread and share information about needs, events, and relief operations. However, these tweets are not easily aggregated into meaningful topics for ease of delivery to people who critically need the information.


Objective

The goal of this research project is to develop and deploy, in collaboration with the NGO consortium NetHope, a reusable text message-based infrastructure that classifies and aggregates multi-lingual tweets and text messages about Haiti relief operations by topics and regions so that they can be easily subscribed to by NGO's, Haitians, and their friends and families.


Synopsis

This research project leverages existing software developed for entity extraction and topic classification so that a system can be developed and deployed quickly to respond to the time critical needs in Haiti. The evaluation of the system includes feedback from NetHope as well as quantitative metrics about its usage and performance. The intellectual merit of this research is the development of an integrated engineering solution to automatically identify topics and geo-locations from brief messages by utilizing contextual information. Such a system can become a foundation for designing next generation information infrastructure for emergency response and disaster management.

College of Information Sciences and Technology ©2010

This material is based upon work supported by the National Science Foundation under Grant No. 1026763. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.