Research and Application of Intersection Clustering Algorithm Based on PCA Feature Extraction and K-Means

Traffic jam and traffic route design are important issues of people's livelihood. In this paper, machine learning algorithm is used to cluster intersection features. Based on the collected traffic intersection data, PCA dimension reduction technology is used for feature extraction, and then K-Means clustering algorithm is used to sum up the number of entrances, and each intersection is clustered. Thus, a new intersection clustering algorithm is established, studied and applied. The algorithm can match similar intersections, which provides theoretical support for the transportation department and related technical personnel.