<![CDATA[Fluctuations in chili prices as a strategic commodity often pose challenges to food price stability, particularly in Central Java regions such as Banyumas and Cilacap Regencies. This study aims to explore and compare the performance of three classical machine learning algorithms: Ridge Regression, Lasso Regression, and Random Forest Regressor. This study utilizes daily historical data spanning from January 2018 to June 2025 to identify market patterns and trends. Furthermore, this research projects chili prices for the next 365 days. Consistently, Lasso Regression demonstrated the best performance in both regions. The model achieved the highest accuracy in Banyumas Regency Banyumas (MAPE 2,75%) and Cilacap Regency (MAPE 5,08%). However, visual analysis of long-term predictions revealed that Ridge Regression produced a more realistic graph compared to Lasso Regression. Conversely, Random Forest Regressor failed to capture long-term trends as it yielded stable predictions. The prediction results were subsequently visualized in an interactive dashboard based on the Flask framework.]]>
Copyrights © 2026
