Klasifikasi Kualitas Udara Menggunakan Algoritma K-Nearest Neighbor Berdasarkan Data Pm2.5 Di Indonesia Tahun 2025

Mukti, Khaila Mukti Harahap
Sofinah Harahap, Lailan

Jurnal Ilmiah Ilmu Komputer dan Teknologi Informasi (JIKOMTI)
Vol. 2 No. 2 (2025): JIKOMTI: DESEMBER 2025

This study aims to classify air quality in various cities in Indonesia based on PM2.5 concentrations using the K-Nearest Neighbor (KNN) algorithm. The data used is sourced from the June–July 2025 edition of Nafas Indonesia, which contains the average PM2.5 values from 21 major cities. The analysis process includes normalization, category labeling based on US EPA standards, model training, and accuracy evaluation. The results show that the KNN model with a k value of 3 is capable of classifying air quality with an accuracy rate of 85%. These findings indicate that the KNN algorithm is quite effective in grouping air quality levels based on particulate pollutant (PM2.5) data.