| dc.description.abstract | Food security remains a crucial issue in development, both globally and nationally. According to data from the annual report of the Directorate General of Food Crops, food production in Kalimantan has shown a declining trend in recent years, which may affect food availability in the Kalimantan region in the future. Although the Indonesian government has developed the food estate program, its implementation has encountered various conflicts that have led to failures. Therefore, precise and structured planning is needed to prevent potential negative impacts. To enhance food production in Kalimantan, it is essential to understand the factors that influence production volume, one of which is harvest area. This study aims to analyze the correlation between harvest area and food production, as well as to compare the performance of the Linear Regression and Random Forest algorithms in predicting food production in Kalimantan. The research data was obtained from the official website of the Ministry of Agriculture of the Republic of Indonesia, consisting of time-series data on harvest area and production of rice and secondary crops from 1970 to 2023. The correlation analysis results indicate that harvest area significantly affects food production, and the Random Forest model produces lower MAE and RMSE values and higher R² values compared to the Linear Regression model. This indicates that the Random Forest model performs better in predicting food production in Kalimantan. This model is then integrated into a web-based application that displays the prediction results, thereby providing insights for stakeholders in strategic planning to prevent food insecurity due to unstable food availability in Kalimantan. | en_US |
