Content area
Full text
Introduction
In the modern day, social networking sites offer a great way to explore the interests, hobbies, and behaviors of user groups. Individuals are posting comments and multimedia information concerning personal lifestyles, likes, sentiments, and opinions more than ever before Vigna et al. (2015). On the other hand, as a result of the widespread usage of social media sites during times of crisis brought on by natural or man-made disasters, there are countless opportunities for individuals seeking information to quickly acquire insightful data Alam et al. (2017). Thus, social media can be used to forecast and identify disasters by looking at people’s posts on social media platforms Jaeger et al. (2007). Using these platforms to monitor real-time updates, images, and comments, authorities and researchers can spot early signs of crises, like unusual weather conditions, changes in sentiment, or localized emergencies. By combining this information, we’ll be able to provide early warnings and respond fast to emerging threats. One of the giant social media sites, Twitter has been used to spread news, assist in the rapid reaction to disasters, and monitor relief and recovery activities. For instance, tweets about the Virginia earthquake spread around the US more quickly than the real earthquake, according to data visualizations of this phenomenon Lotan (2011). Researchers solely focused on the creation of effective technologies like Artificial Intelligence Shaukat et al. (2020a), Time series analysis Shaukat et al. (2021a), Internet Of Things (Shaukat et al. 2021b, 2017; Tariq et al. 2023) to harness and use current data from social networking sites with the goal of humanitarian responses to help relief operations to catastrophes Castillo (2016). They created techniques like automatic information extraction from posts Yin et al. (2012), timely detection of events Ashktorab et al. (2014a), and automated image classification Alam et al. (2018). For example, the Australian Red Cross uses a computer filtration system for spam and the classification of social media posts into event types. Whereas ResilienceDirect helps to collaborate with all UK emergency services by assimilating evidence gathered from social media. Furthermore, the American Red Cross employs a monitoring system that tracks possible emergencies Pekar et al. (2020). Though the rescue request is a time- and resource-consuming process, machine learning (ML) Rabbi et al. (2024a), natural language processing...