Image of EVALUASI METODE UNSUPERVISED LEARNING DALAM PENGELOMPOKAN DATA KECELAKAAN LALU LINTAS UNTUK MENGIDENTIFIKASI POLA SPASIAL BERDASARKAN LOKASI DAN TINGKAT KEPARAHAN
Bookmark Share

Skripsi

EVALUASI METODE UNSUPERVISED LEARNING DALAM PENGELOMPOKAN DATA KECELAKAAN LALU LINTAS UNTUK MENGIDENTIFIKASI POLA SPASIAL BERDASARKAN LOKASI DAN TINGKAT KEPARAHAN

Traffic accidents are one of the transportation issues that require distribution analysis to identify areas with different accident characteristics. This study aims to compare the K-Means, DBSCAN, and Hierarchical Clustering methods in clustering traffic accident data based on geographical location and accident severity. The dataset used is derived from traffic accident data in the United Kingdom (UK) with three features, namely longitude, latitude, and collision severity. The evaluation was conducted using the Silhouette Score and Davies–Bouldin Index (DBI), supported by spatial analysis thru centroids, Haversine distance, and cluster distribution visualization. The research results show that K-Means produces 7 clusters with a Silhouette Score of 0.5280 and a DBI of 0.7742. DBSCAN produces 16 clusters and 130 noise data points with a Silhouette Score of 0.3535 and a DBI of 0.5740, while Hierarchical Clustering produces 10 clusters with a Silhouette Score of 0.4777 and a DBI of 0.8471. Based on the results of the evaluation and spatial interpretation, K-Means provides the best performance because it produces more balanced clusters, is easy to interpret, and can represent the entire data without noise. Therefore, K-Means is considered the most suitable for illustrating the distribution pattern of traffic accidents based on location and severity of accidents in the dataset used.

Availability
#
Central Library (Reference) T2005732026
T200573
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T2005732026
Publisher
Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xvi, 109 hlm.; ilus.; tab.; 29 cm.
Language
Indonesia
ISBN/ISSN
-
Classification
006.307
Content Type
Text
Media Type
-
Carrier Type
-
Edition
-
Subject(s)
Prodi Sistem Komputer
Model Deep Learning
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment