A Comparative Analysis of the Machine Learning Model for Crop Yield Prediction in Quezon Province, Philippines

Digital Object Identifier (DOI)



Pitz Gerald G. Lagrazon

Institute of Graduate Studies and Research Manuel S. Enverga University Foundation

Dr. Jose B. Tan Jr.

Institute of Graduate Studies and Research Manuel S. Enverga University Foundation


The Philippines is predominantly an agriculturally dependent nation; of the 30 million hectares of land, around one-third is classified as agricultural land. In the last 15 years, agriculture has generated roughly 20% of the nation’s Gross Domestic Product, 24% of its entire export revenue, and 46% of its total employment. The Department of Agriculture claims that the restoration of irrigation facilities, favorable weather, improved fertilization, and high-quality seeds contribute to higher crop yield production. However, according to Philippine statistics, the harvest in 2021 decreased by 2.6 percent due to unfavorable weather conditions. This study aims to determine which among the machine learning models such as Support Vector Machine, Decision Tree, Gaussian Process Regression, Ensemble, and Neural Network are the most accurate for crop yield prediction by having the comparative analysis using machine learning algorithms. Machine learning is an important tool for crop yield prediction because it enables computers to learn from data, extract patterns, and make predictions without being explicitly programmed. The datasets were trained and tested to analyze the best algorithm for predicting crop yield. After tuning the hyperparameters of different models, it shows that the GPR outperformed all other models with an RMSE of 0.046579. This will help farmers set goals, evaluate alternatives, specify management plans during the crop production planning process, and optimize their farming practices, giving them a better understanding of their crops and how to maximize them.


Support vector machines
Analytical models
Machine learning algorithms
Computational modeling
Neural networks
Machine learning