Year: 2025
Type: article
Abstract: Natural disasters are characterized as a combination of natural hazards and vulnerabilities that endanger communities and result in significant financial and human losses. The uncertain occurrence of these events and the limited resources in affected regions constitute their core attributes. Leveraging information technology to assess, forecast, and depict these occurrences can improve handling such disasters. Artificial Intelligence (AI) technology has recently been employed practically in various disciplines for various applications, and disaster management applications are among the most critical areas. The dataset is essential to any AI system and must be corrected to deliver reliable results. Disaster datasets are presented and described in literature. This study examines the datasets available for disaster evaluation and detection frameworks. A general framework consisting of five main dataset properties is designed for the presentation of the datasets. Some features include general information, data volume, recording environment, data nature, and evaluation. There are two or three sub-properties under each of the five main properties. Each dataset is examined in comparison to the listed attributes.
Authors: Sohail Abbas, Manar Abu Talib, Qasim Nasir, Omar Belal, Mohammed Emad Al-Haidary, Ibtihal Ahmed
Institution: University of Sharjah
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $1,134
Funder: University of Sharjah
Award: G4272520181
OpenAlex ID: W4410896985