Optimasi Algoritma K-Nearest Neighbor Menggunakan Particle Swarm Optimization untuk Prediksi Penyakit Cacar Monyet

Monkeypox is a zoonotic disease that can spread through direct contact with infected humans or animals. The increasing number of cases reported in various countries highlights the importance of early detection to support timely diagnosis and treatment. One promising approach is the application of Machine Learning techniques, particularly the K-Nearest Neighbor (KNN) algorithm. However, the performance of KNN is highly dependent on the selection of the optimal K parameter. Therefore, an optimization method is required to improve its classification performance. This study aims to optimize the K-Nearest Neighbor algorithm using Particle Swarm Optimization (PSO) for monkeypox disease prediction. The dataset used consists of 500 records with 50 attributes obtained from Kaggle. The research methodology includes Exploratory Data Analysis (EDA), data preprocessing, training and testing data partitioning, baseline KNN model development, optimization of the K parameter using PSO, and model evaluation using Accuracy, Precision, Recall, F1-Score, and Confusion Matrix metrics. The results indicate that PSO optimization improves the performance of the KNN model by identifying a more optimal K value compared with the baseline model. These findings demonstrate that the integration of PSO and KNN has the potential to enhance classification performance and can be considered an effective approach for developing decision support systems for the early detection of monkeypox disease.