PREDIKSI TINGKAT KUALITAS UDARA DENGAN PENDEKATAN ALGORITMA K-NEAREST NEIGHBOR

Abstract

Air is the source of life for living things. However, in recent years the decline in air quality has become a serious problem that urgently needs to be addressed. The Air Pollution Standard Index (ISPU) is an indicator that can be used to determine the condition of ambient air quality in a particular location. Yogyakarta is one of the major cities in Indonesia with serious air pollution problems in recent years. This research aims to predict air quality in Yogyakarta City based on ISPU data using data mining techniques and classification methods. The algorithm used in prediction is K-Nearest Neighbor (K-NN), which classifies new objects based on their nearest neighbors. Evaluation of the algorithm model is done by measuring accuracy, precision, recall, and f-measure for the value of K = 5. The test results show that the value of K = 5 provides good performance with an accuracy of 99% which is for the "Good" category producing 100% precision, 99% recall, and 100% f-measure, while the "Moderate" category produces 98% precision, 100% recall, and 99% f-measure.