Predicting Crop Yield in Quezon Province, Philippines Using Gaussian Process Regression: A Data-Driven Approach for Agriculture Sustainability

Digital Object Identifier (DOI)



Pitz Gerald G. Lagrazon

Institute of Graduate Studies and Research

Jose B. Tan

Institute of Graduate Studies and Research


This study presents a predictive application for rice and corn crop yields in Quezon Province, Philippines, using advanced machine learning techniques, with a focus on the Gaussian Process Regression model. The desktop application utilizes weather parameters as inputs to forecast crop volumes, offering farmers valuable insights for optimized planting and harvesting decisions. Through rigorous evaluation, the Gaussian Process Regression model consistently outperforms other models, demonstrating its accuracy and potential for practical use in the agricultural sector. With an overall Mean Absolute Percentage Error (MAPE) of 3.39%, this tool holds promise for enhancing food security and sustainable agriculture while serving as a model for similar initiatives in other regions. Furthermore, this research highlights the crucial role of data-driven approaches in addressing climate change impacts on agriculture, offering a tangible solution for a more food-secure future. The study underscores the wider potential of data science and machine learning in addressing pressing global concerns, promoting more resilient and sustainable agricultural practices and ensuring a food-secure future for all.