基于数据融合的高速公路用户画像及聚类分析OA

To address the challenges of complex and diverse travel behaviors of freight vehicles in Tianjin and the lack of clear user identification,a comprehensive dataset integrating data from the Tianjin Expressway system and a freight logistics platform was constructed,containing multi-dimensional attributes including vehicle trajecto-ry,vehicle type,and registration location.Data cleaning,matching,and integration were performed on the Python platform.The K-means clustering algorithm was applied for unsupervised learning on high-dimension-al feature vectors,and a relatively high K value was set to enable fine-grained travel pattern recognition.This approach effectively reveals the travel diver.