Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Bulletin of Network Engineer and Informatics (BUFNETS)
Wawan Joko Pranoto
Department of Informatics Engineering, Universitas Muhammadiyah Kalimantan Timur, Indonesia
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Application of the Forward Selection and ADASYN Methods in the K-Nearest Neighbor (KNN) Algorithm for Traffic Accident Data Classification Agus Tri Waluyo; Taghfirul Azhima Yoga Siswa; Wawan Joko Pranoto
Bulletin of Network Engineer and Informatics Vol. 4 No. 2 (2026): BUFNETS (Bulletin of Network Engineer and Informatics) October 2026
Publisher : PT. GWEX NET PUBLISHER

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59688/px197639

Abstract

Traffic accidents are a serious problem that threaten public safety and cause material losses, particularly in Samarinda City, where the number of cases rose by 69.61% between 2021 and 2023. This study aims to improve the performance of a traffic-accident-severity classification model by optimizing the K-Nearest Neighbor (KNN) algorithm through a combination of Adaptive Synthetic Sampling (ADASYN) and Forward Selection. The dataset, obtained from accident reports of the Samarinda City Police for 2020–2024, consists of 2,385 records and 35 attributes, split using 10-fold cross-validation and evaluated with a confusion matrix. The standard KNN model achieved an accuracy of 79.61%, precision of 25.06%, recall of 23.81%, and an f1-score of 24.29%. After applying ADASYN, performance improved to an accuracy of 92.54%, precision of 93.08%, recall of 92.01%, and an f1-score of 92.50%. The largest gain occurred after Forward Selection narrowed the feature set to four attributes, yielding an accuracy of 99.62%, precision of 99.89%, recall of 99.34%, and an f1-score of 99.62%. These results demonstrate that combining ADASYN with Forward Selection significantly improves the classification performance of the KNN model in predicting traffic-accident severity, and offers an effective approach for future accident-severity prediction systems.
Co-Authors Agus Tri Waluyo Taghfirul Azhima Yoga Siswa