AN ANALYSIS BASED ON MULTI CRITERIA STRATEGIES USING MACHINE LEARNING AND STATISTICS FOR PULSES CROP YIELD ANALYSIS IN TAMIL NADU REGION

AN ANALYSIS BASED ON MULTI CRITERIA STRATEGIES USING MACHINE LEARNING AND STATISTICS FOR PULSES CROP YIELD ANALYSIS IN TAMIL NADU REGION

Agriculture plays a crucial role in India's economic expansion. However, increasing population and a continuously changing climate significantly impact crop productivity and national food security. The agricultural sector operates within a complex system, generating vast amounts of data from multiple variables. Machine learning, when applied dynamically and efficiently, has the potential to enhance agricultural decision-making by predicting crop yields, selecting optimal crops for cultivation, and determining the best treatment strategies throughout the growing season. Various approaches exist for predicting crop production using environmental variables, and this research investigates some of these methods. The primary objective of this paper is to develop a machine learning model alongside a statistical approach for yield prediction. In addition to improving efficiency, machine learning algorithms and statistical techniques can assist farmers in selecting the most suitable crops by considering multiple influencing factors. This study explores the effectiveness of several machine learning methods in agricultural applications. The techniques examined include elastic net regression, ridge regression, least absolute shrinkage and selection operator (LASSO) regression, artificial neural networks (ANN), and extreme gradient boosting (XG Boost). The performance of these models was evaluated using R-squared (R²), mean absolute error (MAE), root mean squared error (RMSE), mean squared error (MSE), and mean bias deviation (MBD).