<![CDATA[Rice holds importance in Indonesia as a commodity driving the economy and improving societal well-being, however, its production encounters obstacles attributed to the effects of drastic climate variations. This study sought to evaluate how Support Vector Regression (SVR) k Nearest Neighbors (KNN) and Decision Tree models perform in forecasting rice yields while considering variables related to climate change. The research process included stages such, as gathering and cleaning the information before exploring and analyzing it to apply metrics and implement algorithms like Mean Absolute Error (MAE) Root Mean Squared Error (RMSE) and R² Score, for evaluation purposes. The findings obtained from the study indicate that the Decision Tree technique is efficient, achieving a minimal deviation rate of 0%. This outcome implies that the model effectively grasped the core patterns within the dataset while reducing errors effectively. The KNN model displayed performance levels and suggested room, for enhancement with parameter adjustments; however, SVM Regression was deemed fitting for the datasets needs. The results emphasize the significance of choosing the algorithm for modeling in agriculture and stress the necessity, for additional research to confirm these findings in various datasets.]]>
Copyrights © 2025
