CD Skripsi
Analisis Pengelompokan Penyebab Laka Lantas Di Kota Pekanbaru Menggunakan Algoritma K-Means
Traffic accidents are a serious issue in Pekanbaru City, affecting public safety and well-being. This study aims to classify the causes of traffic accidents using the K- Means Clustering algorithm based on accident data from Riau Regional Police (POLDA Riau) for the period 2021–2023. The clustering process uses Euclidean distance to measure the closeness of data points to each cluster’s centroid. The optimal number of clusters was determined using the Elbow Method, which indicated K=2 as the best value. Clustering quality was evaluated using the Silhouette Index, resulting in a score of 0.293, which indicates a reasonably good grouping. The analysis shows that Cluster 2 has a higher number of accidents than Cluster 1 across various aspects such as victim age, vehicle type, accident cause, and time of occurrence. Most accidents in Cluster 2 involved 19-year-old victims, were caused by speeding, and occurred between 12:00–18:00 WIB. These findings are expected to support POLDA Riau in formulating more effective accident prevention policies.
Keywords: Clustering, K-Means, Pekanbaru, POLDA Riau, Traffic Accidents
Tidak tersedia versi lain