1.1
1
xml
info:srw/schema/1/mods-v3.2
Analysis of Detection Models for Disaster-Related Tweets
Matti Wiegmann
author
Jens Kersten
author
Friederike Klan
author
Martin Potthast
author
Benno Stein
author
2020
Virginia Tech
Blacksburg, VA (USA)
English
Social media is perceived as a rich resource for disaster management and relief efforts, but the high class imbalance between disaster-related and non-disaster-related messages challenges a reliable detection. We analyze and compare the effectiveness of three state-of-the-art machine learning models for detecting disaster-related tweets. In this regard we introduce the Disaster Tweet Corpus~2020, an extended compilation of existing resources, which comprises a total of 123,166 tweets from 46~disasters covering 9~disaster types. Our findings from a large experiments series include: detection models work equally well over a broad range of disaster types when being trained for the respective type, a domain transfer across disaster types leads to unacceptable performance drops, or, similarly, type-agnostic classification models behave more robust at a lower effectiveness level. Altogether, the average misclassification rate of~3,8\% on performance-optimized detection models indicates effective classification knowledge but comes at the price of insufficient generalizability.
Tweet Filtering
Crisis Management
Evaluation Framework
matti.wiegmann@uni-weimar.de
exported from refbase (http://idl.iscram.org/show.php?record=2278), last updated on Mon, 29 Jun 2020 07:53:51 +0200
text
http://idl.iscram.org/files/mattiwiegmann/2020/2278_MattiWiegmann_etal2020.pdf
MattiWiegmann_etal2020
ISCRAM 2020 Conference Proceedings – 17th International Conference on Information Systems for Crisis Response and Management
Iscram 2020
Amanda Hughes
editor
Fiona McNeill
editor
Christopher W. Zobel
editor
17th International Conference on Information Systems for Crisis Response and Management
2020
Virginia Tech
Blacksburg, VA (USA)
conference publication
872
880
2411-3463
978-1-949373-27-77
1