Prediction of Early Floods Under Indecisive Weather Forecasts Using Innovative K Nearest Neighbor

One of the most damaging natural disasters, floods has a major influence on the environment and society by submerging communities and upsetting ecosystems. Early flood prediction is essential because it protects communities, minimizes possible damage, and allows for prompt responses. The work uses sophisticated machine learning techniques like K-Nearest Neighbors (KNN) and Gradient Boosting Machine (GBM) to estimate flood events using a historical flood and rainfall dataset from Kaggle, a well-known platform for collaborative machine learning and data science projects. The results, which were calculated using Python, demonstrate that KNN outperformed GBM, which had an accuracy of 90%, with an accuracy of 98% when 40 dataset samples were analyzed with KNN and 20 samples with GBM. The significance value, which indicates statistical significance, was found to be <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{p}=0.000(\mathrm{p} &lt; 0.05)$</tex> following an independent samples T-Test study. The results demonstrate KNN's potential as a formidable tool for improving disaster preparedness plans in addition to demonstrating the accuracy of KNN and GBM in early flood prediction. The use of machine learning for precise flood forecasting has advanced significantly as a result of this work.