Innovations in Geospatial Data Analysis: Applied Nonlinear Analysis, Remote Sensing, AI, and GIS for Environmental Sustainability

Abstract

Abstract: 
Water mapping plays a pivotal role in sustainable water resource management, particularly in the context of escalating climate change impacts. This study addresses the critical need for accurate water level predictions by introducing an innovative ensemble machine learning (ML) approach. Motivated by the increasing importance of ML and mathematical models in geospatial data analysis, the research papers from the diverse database, revealing a research gap that includes the absence of standardized methodologies and the exploration of diverse ensemble methods. Leveraging the EuroSat dataset for land use and land cover classification, the proposed ensemble model combines Principal Component Analysis, Genetic Algorithm, Gradient Boosted Decision Trees, Frequency Ratio, Deep Neural Network, and Shannon's Entropy. Demonstrating superior accuracy at 98.5%, Precision at 90.4%, and recall at 92%, EnsembleML emerges as a robust solution, emphasizing the advantage of ensemble techniques for comprehensive water mapping in the face of environmental challenges.