Paper Title
MACHINE LEARNING-BASED SOIL CLASSIFICATION USING NEURAL NETWORKS AND GRADIENT BOOSTING USING PENINSULAR MALAYSIA IN SITU DATA

Abstract
The classification of soil is an important factor in assessing a piece of land's viability for different types of agriculture. In Peninsular Malaysia, this study uses machine learning approaches to categorize different types of soil and suggest appropriate crops based on soil attributes. This study adopted two machine learning models using data from 23 locations: a 97% accuracy model using neural networks and a 91% accuracy model using gradient boosting classifiers. We also included crop recommendations to help farmers make well-informed choices. This study highlights how machine learning may enhance land management and agricultural productivity. Keywords - Gradient Booster Classifier, Soil Classification, Machine Learning, Neural Network (NN), Streamlit, and Crop Recommendation, Agriculture,.