Ejup Rustemi1*
1 University of Tetova, North Macedonia
https://orcid.org/0009-0003-5133-7834
Mefail Tahiri2
2University of Tetova, North Macedonia
https://orcid.org/0009-0007-5117-4560
Corresponding Author: * ejup.rustemi@yahoo.com
Abstract
Over the course of the past few years, technology known as artificial intelligence (AI) has developed as a strong tool in a variety of sectors. The application of machine learning techniques has become widespread in fields such as medicine and finance as a result of advancements in algorithmic design, an increase in the amount of computer power available, and the availability of an excessive amount of data. On the other hand, there has been a recent uptick in interest about the utilization of AI approaches for the purpose of minimizing the destructive effects of natural disasters. Not only does artificial intelligence offer a way to strengthen natural disaster management during the ongoing event, but it also offers a way to alleviate some of the mental health conditions and support the psychosocial well-being of affected citizens, survivors, and exposed responders. This is because AI has the potential to enhance prediction, response, and recovery efforts. This latter point is particularly intriguing in light of the ever-increasing prevalence of mental health conditions across the globe, particularly in the context of growing climate crises, where the requirement for interventions that are both accessible and effective continues to be extremely relevant.
Keywords: AI, disaster, management, technology, machine learning.
Introduction
In response to the emergence of disruptive technologies such as artificial intelligence, new opportunities to improve catastrophe resilience and risk reduction are being explored. In the monitoring of earthquakes and floods, models driven by artificial intelligence are rapidly proving to be successful. An example of this would be the M-LARGE project, in which researchers merged satellite photos with deep learning models in order to provide an early warning signal prior to the commencement of significant shaking and tsunamis. With this invention, conventional methods were routinely surpassed, which resulted in improved predictions in areas that had been affected earlier. An additional illustration of this is the application of neural networks in urban flood nowcasting, which is the process of predicting the state of floods in the near future in response to the unfolding of catastrophic weather. For the purpose of modeling and predicting the progression of a flooding catastrophe, the researchers utilized community aspects such as tweets, emergency calls, and other such data. As a result, the community is better able to foresee dangers and repercussions, and it can also assist bring about situational awareness.