Analisis Algoritma C45 dan Regresi Linear untuk Memprediksi Hasil Panen Kelapa Sawit

Indonesia, as one of the main producers of palm oil in the world, has an agricultural sector that is very influential on the national economy, especially through palm oil exports. Prediction of oil palm yields is crucial to improve efficiency in planning and resource management. This study was conducted to compare the performance of two prediction methods, namely the C45 Algorithm and Linear Regression, in predicting oil palm yields. The formulation of the problems raised in this study includes: (1) How does the performance of the C45 Algorithm and Linear Regression compare in predicting oil palm yields? (2) How accurate are the predictions generated by the two algorithms based on historical data on crop yields? (3) What are the factors that influence the choice between C45 Algorithm and Linear Regression for oil palm yield prediction? The data used in this study is historical data from PT. Surya Inti Sawit Kahuripan, which includes 106 data blocks. The variables analyzed included land area, number of trees, number of trees per hectare, planting year, soil type, fertilizer use plan, and yield in tons. Data analysis was carried out using the C45 Algorithm, which forms a decision tree based on historical data, and the Linear Regression method, which analyzes the linear relationship between independent variables and dependent variables. Prediction accuracy is measured using Root Mean Squared Error (RMSE). The results show that the C45 Algorithm has a lower RMSE value than Linear Regression, indicating that the C45 Algorithm provides more accurate predictions.