Predictive Data Analysis to Support Decision-Making Based on Long-Term Impacts of Disasters

Research into disaster relief is a field with a lot of potential, and predictions made about its future are based on historical data. It is a difficult task to analyze the long-term consequences of a disaster. Governments, NGOs, and other organizations struggle to make sense of disaster data. It is hard to assess the impact of different factors, such as education, employment, agriculture, and the social sector, on people living in affected areas. This research paper represents a framework to evaluate the long-term effects of disasters, by means of machine learning algorithms. The algorithms mentioned in this paper are Random Forest, Decision Tree, Gradient-boosted Decision Tree (GBDT), and XGBoost. It seeks to analyze historical disaster data, identify impacts, and predict the effects of similar disasters in the future. The framework also suggests preemptive measures and the impact on those involved. Finally, the proposed framework will be able to make long-term analyses and recommendations, using data from past events.