Indonesia is one of the countries that often experiences abnormal weather cycles. Extreme weather occurs due to several factors, such as unusual atmosphere. Weather prediction is very important in helping people anticipate various possibilities caused by extreme weather changes. This study aims to classify the weather in Bandung using the K-Nearest Neighbor algorithm. The K-Nearest Neighbor algorithm was chosen because of its ty to handle data with not too many variables and its implementation in weather classification cases. The dataset is secondary data taken from kaggle.com. Testing was conducted with k = 30, k = 50 and k = 70. With the highest accuracy, precision, and recall values at k = 50.
Copyrights © 2025