A Hybrid Machine Learning Approach for Crop Classification, Yield and Fertilizer Prediction for Sustainable Agriculture

Nakagiri, Lynnet Grace (2024) A Hybrid Machine Learning Approach for Crop Classification, Yield and Fertilizer Prediction for Sustainable Agriculture. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (697kB) | Preview
Preview	PDF (Configuration Manual) Download (393kB) | Preview

Abstract

The increase in world population has put pressure on food production and this has made it more crucial to make sustainable decision-making in agriculture. This study proposed a hybrid machine learning model to predict crops, yield, and fertilizer requirements. Base models (Random Forest, Gradient Boosting, and LSTM) are integrated with meta-models (SVR, XGBoost, and MLP). MLP meta-model was the best hybrid framework with accuracy of 99.96% and RMSE of 0.0415 for crop classification. RMSE and MAE of 0.1628 and 0.0912 were achieved in yield prediction and RMSE of 0.0294 and R² of 0.9990 in fertilizer prediction. These results highlight the hybrid model's capacity to address agricultural challenges and its potential to enhance precision agriculture technologies. This shows that hybrid model is capable of addressing agricultural challenges.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Nagahamulla, Harshani UNSPECIFIED
Uncontrolled Keywords:	Machine learning; agriculture; crop classification; yield prediction; fertilizer optimization; hybrid model; precision agriculture
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science S Agriculture > S Agriculture (General) Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	03 Sep 2025 14:51
Last Modified:	03 Sep 2025 14:51
URI:	https://norma.ncirl.ie/id/eprint/8756

Actions (login required)

View Item