Social-media text mining and network analysis to support decision support for natural crisis management
Social-media text mining and network analysis to support decision support for natural crisis management
A core issue in crisis management is to extract from the mass of incoming information what is important for situational awareness during mass emergencies. Based on a case study we develop a prototypical application, TweetComp1, which is integrated into the decision-support component of a Tsunami early warning system and demonstrates the applicability of our approach. This paper describes four novel approaches using focused twitter crawling, trustworthiness analysis, location analysis/geo-parsing, and multilingual tweet classification in the context of how they could be used for monitoring crises. The validity of our state-of-the art text mining and network analysis technologies will be verified in different experiments based on a human annotated gold standard corpus.
Middleton, Stuart
404b62ba-d77e-476b-9775-32645b04473f
Zielinski, Andrea
76ccfc08-52b2-408a-ac49-5c306e7c0bd0
Tokarchuk, Laurissa N.
a4133477-9309-44fb-96fa-414e19f528f3
Wang, Xinyne
930e26d2-8a0a-4676-a861-d0bb6f95ee1f
May 2013
Middleton, Stuart
404b62ba-d77e-476b-9775-32645b04473f
Zielinski, Andrea
76ccfc08-52b2-408a-ac49-5c306e7c0bd0
Tokarchuk, Laurissa N.
a4133477-9309-44fb-96fa-414e19f528f3
Wang, Xinyne
930e26d2-8a0a-4676-a861-d0bb6f95ee1f
Middleton, Stuart, Zielinski, Andrea, Tokarchuk, Laurissa N. and Wang, Xinyne
(2013)
Social-media text mining and network analysis to support decision support for natural crisis management.
ISCRAM 2013, Baden-Baden, Germany.
12 - 15 May 2013.
Record type:
Conference or Workshop Item
(Paper)
Abstract
A core issue in crisis management is to extract from the mass of incoming information what is important for situational awareness during mass emergencies. Based on a case study we develop a prototypical application, TweetComp1, which is integrated into the decision-support component of a Tsunami early warning system and demonstrates the applicability of our approach. This paper describes four novel approaches using focused twitter crawling, trustworthiness analysis, location analysis/geo-parsing, and multilingual tweet classification in the context of how they could be used for monitoring crises. The validity of our state-of-the art text mining and network analysis technologies will be verified in different experiments based on a human annotated gold standard corpus.
More information
Published date: May 2013
Venue - Dates:
ISCRAM 2013, Baden-Baden, Germany, 2013-05-12 - 2013-05-15
Organisations:
IT Innovation
Identifiers
Local EPrints ID: 359364
URI: http://eprints.soton.ac.uk/id/eprint/359364
PURE UUID: 68fb9998-c706-469b-a0dc-393d98c0385e
Catalogue record
Date deposited: 30 Oct 2013 14:22
Last modified: 15 Mar 2024 03:08
Export record
Contributors
Author:
Andrea Zielinski
Author:
Laurissa N. Tokarchuk
Author:
Xinyne Wang
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
