A multi-level analytic framework for disaster situational awareness using Twitter data
Abstract During a natural disaster, mining messages from social media platforms can facilitate local agencies, rescue teams, humanitarian aid organizations, etc., to track the situational awareness of the public. However, for different stakeholders, the concerns about people’s situational awareness...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2022-08-01
|
| Series: | Computational Urban Science |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43762-022-00052-z |
| _version_ | 1811215537823285248 |
|---|---|
| author | Wei Zhai |
| author_facet | Wei Zhai |
| author_sort | Wei Zhai |
| collection | DOAJ |
| description | Abstract During a natural disaster, mining messages from social media platforms can facilitate local agencies, rescue teams, humanitarian aid organizations, etc., to track the situational awareness of the public. However, for different stakeholders, the concerns about people’s situational awareness in a natural disaster event are different. Therefore, I developed a Twitter-based analytic framework to take perception-level situational awareness, humanitarian-level situational awareness, and action-level situational awareness into consideration. Specifically, perception-level situational awareness mainly reflects people’s perception of the ongoing natural disaster event (i.e., if people are discussing the disaster event). Decision-makers can rapidly have a big picture of severely impacted regions. Humanitarian-level situational awareness represents tweets that are associated with the humanitarian categories based on the definition from the United Nations Office for the Coordination of Humanitarian Affairs. The detection of humanitarian-level situational awareness can help response teams understand the specific situations and needs of local communities. In terms of the action-level situational awareness, I extracted noun-verb pairs in each tweet to explicitly represent the specific event described in a given tweet, so that the response teams can quickly act on the situation case by case. Moreover, to shed light on disaster resilience and social vulnerability, I further examined the demographic characteristics of three levels of situational awareness. I empirically demonstrated the analytic framework using geo-tagged tweets during 2018 Hurricane Michael. |
| first_indexed | 2024-04-12T06:24:10Z |
| format | Article |
| id | doaj.art-e99cc7dceed94018ac6e1434b0b632f1 |
| institution | Directory Open Access Journal |
| issn | 2730-6852 |
| language | English |
| last_indexed | 2024-04-12T06:24:10Z |
| publishDate | 2022-08-01 |
| publisher | Springer |
| record_format | Article |
| series | Computational Urban Science |
| spelling | doaj.art-e99cc7dceed94018ac6e1434b0b632f12022-12-22T03:44:13ZengSpringerComputational Urban Science2730-68522022-08-012111510.1007/s43762-022-00052-zA multi-level analytic framework for disaster situational awareness using Twitter dataWei Zhai0The University of Texas at San AntonioAbstract During a natural disaster, mining messages from social media platforms can facilitate local agencies, rescue teams, humanitarian aid organizations, etc., to track the situational awareness of the public. However, for different stakeholders, the concerns about people’s situational awareness in a natural disaster event are different. Therefore, I developed a Twitter-based analytic framework to take perception-level situational awareness, humanitarian-level situational awareness, and action-level situational awareness into consideration. Specifically, perception-level situational awareness mainly reflects people’s perception of the ongoing natural disaster event (i.e., if people are discussing the disaster event). Decision-makers can rapidly have a big picture of severely impacted regions. Humanitarian-level situational awareness represents tweets that are associated with the humanitarian categories based on the definition from the United Nations Office for the Coordination of Humanitarian Affairs. The detection of humanitarian-level situational awareness can help response teams understand the specific situations and needs of local communities. In terms of the action-level situational awareness, I extracted noun-verb pairs in each tweet to explicitly represent the specific event described in a given tweet, so that the response teams can quickly act on the situation case by case. Moreover, to shed light on disaster resilience and social vulnerability, I further examined the demographic characteristics of three levels of situational awareness. I empirically demonstrated the analytic framework using geo-tagged tweets during 2018 Hurricane Michael.https://doi.org/10.1007/s43762-022-00052-zDisaster managementSituational awarenessTwitter dataNatural language processingMachine learning |
| spellingShingle | Wei Zhai A multi-level analytic framework for disaster situational awareness using Twitter data Computational Urban Science Disaster management Situational awareness Twitter data Natural language processing Machine learning |
| title | A multi-level analytic framework for disaster situational awareness using Twitter data |
| title_full | A multi-level analytic framework for disaster situational awareness using Twitter data |
| title_fullStr | A multi-level analytic framework for disaster situational awareness using Twitter data |
| title_full_unstemmed | A multi-level analytic framework for disaster situational awareness using Twitter data |
| title_short | A multi-level analytic framework for disaster situational awareness using Twitter data |
| title_sort | multi level analytic framework for disaster situational awareness using twitter data |
| topic | Disaster management Situational awareness Twitter data Natural language processing Machine learning |
| url | https://doi.org/10.1007/s43762-022-00052-z |
| work_keys_str_mv | AT weizhai amultilevelanalyticframeworkfordisastersituationalawarenessusingtwitterdata AT weizhai multilevelanalyticframeworkfordisastersituationalawarenessusingtwitterdata |