Date of Award
Summer 2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Engineering Management & Systems Engineering
Program/Concentration
Engineering Management and Systems Engineering
Committee Director
Samuel Kovacic
Committee Member
Ghaith Rabadi
Committee Member
Andres Sousa-Poza
Committee Member
Jiang Li
Abstract
Hurricanes pose a significant threat to both human lives and infrastructure. Decision-makers face substantial challenges during such events, as they must act quickly to address victims’ needs. Social media platforms provide a valuable source for quick and real-time information. Recent hurricane events have shown that people turn to social media to call for help when official communication channels, such as 911, are overwhelmed. However, extracting actionable information from the massive number of messages posted on social media is challenging. Furthermore, verifying social media messages posted by the public is a critical concern for disaster response practitioners, making them hesitant to use this information. This study tackles the problem of identifying and assessing the reliability of actionable rescue messages posted on Twitter during hurricanes. A novel deep learning model is proposed for identifying rescue tweets, integrating a fine-tuned BERT model to extract low-level statistical features from the text and rule-based Regex filters to extract problem-specific features. In addition, a rule-based scoring model is introduced to assess the reliability of the identified rescue messages using a set of reliability indicators derived from the literature. The proposed models were evaluated using data collected and annotated from various hurricane events. The results indicated that the proposed classification model for identifying rescue tweets provides more robust results compared to previous classification methods. Evaluated on rescue tweets from Hurricane Harvey, the proposed reliability assessment model could effectively identify reliable rescue tweets. The models developed in this study aim to improve the quality of actionable rescue information extracted from social media during hurricane events, enabling first responders to effectively integrate social media channels as a supplementary source of information in their decision-making process.
Rights
In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).
DOI
10.25777/3d48-h856
ISBN
9798384456100
Recommended Citation
Khallouli, Wael.
"Harnessing Social Media for Disaster Response: Intelligent Identification of Reliable Rescue Requests During Hurricanes"
(2024). Doctor of Philosophy (PhD), Dissertation, Engineering Management & Systems Engineering, Old Dominion University, DOI: 10.25777/3d48-h856
https://digitalcommons.odu.edu/emse_etds/235
ORCID
0000-0003-2542-5454
Included in
Artificial Intelligence and Robotics Commons, Emergency and Disaster Management Commons, Numerical Analysis and Scientific Computing Commons, Systems Engineering Commons