Improving Rice Yield Prediction Accuracy Using Regression Models with Climate Data

Rice production is critical to food security, and accurate yield predictions are required for planning and decision-making. However, precisely predicting rice yields using machine learning models can be difficult due to the complicated interactions of various factors, such as how climate affects rice production. This study sought to solve this rice production is critical to food security, and accurate yield predictions are required for planning and decision-making. However, accurately predicting rice yields using machine learning models can be difficult due to the complicated interactions of various factors, such as how climate affects rice production. This study aims to address this issue by investigating how climate data affect Malaysian rice yield prediction models. The study used a linear regression model trained on rice production data and compared its performance with models incorporating climate data. Both datasets covered the period from 2010 to 2021 in Malaysia. The study found that including climate data significantly improved the prediction accuracy, with an approximately 77% improvement in MAE and 69% in RMSE. The results suggest that incorporating climate data into yield prediction models is essential for accurate and reliable predictions. These findings have important implications for stakeholders in the agricultural industry who can use accurate yield predictions to make informed decisions. However, the study’s limitations include using a single predictive model and data from a single country, suggesting the need for future studies to explore other machine learning algorithms and expand the scope of the research to other regions. Overall, this study contributes to the growing body of literature on the impact of climate data on yield prediction models and highlights the importance of considering climate data in agricultural decision-making.