Comparative Analysis of Various Machine Learning Models for Lentil Yield Prediction

Agriculture plays a pivotal role in India's economy, with a large segment of the population dependent on this sector. As the nation grapples with rapid population growth, ensuring food security becomes increasingly critical, necessitating more accurate crop yield prediction methods. This study focuses on evaluating the effectiveness of various machine learning models for predicting lentil crop yield, using growth attributes as predictors. Traditional approaches like multivariate linear regression (MLR) have shown limitations in accuracy. In contrast, machine learning techniques such as decision trees (DT), support vector machines (SVM), and artificial neural networks (ANN) offer superior alternatives by effectively managing the complex and non-linear interactions between input variables (e.g., soil properties, environmental conditions, and agricultural practices) and yield outcomes. Data from the Agricultural Research Farm at the School of Agriculture, Graphic Era Hill University, Dehradun, were utilized for model training and testing. Model performance was assessed using root mean square error (RMSE) and mean square error (MSE) metrics. The findings highlight that tree-based models and SVMs achieved the highest prediction accuracy, effectively capturing the variability in lentil yield, thus marking a significant advancement over traditional statistical methods.