mrhenree/Flood-susceptibility-forecast-using-machine-learning-algorithms

The risk of floods has naturally increased due to climate change variability, rapid urbanization, and rapidly expanding spatial development. This hazard to human lives and the global economy has become quite severe. The study focused on evaluating the risk of pluvial flooding in Oyo State, Nigeria, and creating susceptibility forecast models. ### Note: A comprehensive breakdown of the project is contained in the "Flood_susceptibility_forecast_using_machine_learning_models-3" paper; the jupyter notebook bears the project code. ## Project brief The study focused on forecasting flood susceptibility, as well as enhancing the management of potential pluvial flooding risk through the application of machine learning techniques. The dataset for this study was gathered from the Ibadan Metropolis in Oyo state, South West Nigeria, by the Copernicus Climate Data Store and the United States Geological Survey (USGS) using the ArcGIS software. A total of 144,401 records and 8 conditioning variables out of 53 were gathered. Various features including drainage, slope, coordinates (X and Y), curvature, rainfall, flow accumulation, aspect, and topographic wetness index were utilized to predict flooding susceptibilities in this study, using two Machine Learning algorithms: LogisticRegression and RandomForest.