A survey of disaster management datasets

作     者：Abbas, Sohail Talib, Manar Abu Nasir, Qasim Belal, Omar Al-Haidary, Mohammed Emad Ahmed, Ibtihal 

作者机构：Univ Sharjah Coll Comp & Informat Dept Comp Sci Sharjah U Arab Emirates Univ Sharjah Coll Comp & Informat Dept Comp Engn Sharjah U Arab Emirates 

出 版 物：《JOURNAL OF INFORMATION AND TELECOMMUNICATION》 (J. Information. Telecommun.)

年 卷 期：2025年

核心收录：

基　　金：University of Sharjah  UAE 

主　　题：Natural disasters Artificial Intelligence datasets properties image dataset social media dataset 

摘      要：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.