Chili Commodity Price Forecasting in Bandung Regency using the Adaptive Synthetic Sampling (ADASYN) and K-Nearest Neighbor (KNN) Algorithms

Abstract

Red chili is an important spice in Indonesia. The Ministry of Agriculture (MoA) reported the significant contribution of red chili to the Indonesian economy both locally and internationally. Chili plants experience price inflation from year to year. This price change is influenced by several factors such as the number of requests and changes in the weather, both of which can affect production. In this study, the prediction of chili prices was carried out using the K-Nearest Neighbor (KNN) algorithm based on chili price data and weather data. The data obtained had imbalanced classes, so the Adaptive Synthetic (ADASYN) algorithm was used to overcome this issue. From the results, the classification using KNN reached the highest accuracy of 93% but with an F1-Score of 0%. In contrast, classification using KNN and ADASYN obtained 100% accuracy and an F1-Score of 100%.