Implementasi K-Nearest Neighbor untuk Prediksi Harga Bibit Kelapa Sawit dengan Metode CRISP-DM

Oil palm is the most productive vegetable oil producing crop and plays an important role in the Indonesian economy. However, fluctuations in the price of oil palm seedlings are a challenge for the Oil Palm Research Center (PPKS) Sungai Lilin Plantation due to the imbalance between demand and production. This study aims to predict the price of oil palm seedlings using the K-Nearest Neighbor (KNN) algorithm with the Cross-Industry Standard Process for Data Mining (CRISP-DM) approach. The dataset used consists of 2,049 records, with a division of 80% training data and 20% test data. With a value of k = 5, the KNN model produces 99.21% accuracy, with 3 misclassified data and 1,626 correct data. The prediction results show a price category of “EXPENSIVE”. This study proves that the KNN method is effective in predicting the price of oil palm seedlings, so it can help stakeholders in decision-making and business strategy