Road network subarea division and evaluation method based on Canopy + Kmeans clustering

Abstract

The invention relates to the field of neural network technical methods, in particular to a road network sub-region division and evaluation method based on Canopy + Kmeans clustering, which takes real-time acquisition of road section center latitude and longitude, road section average speed and road section average density as sample data, and comprises the following specific steps: (1) carrying outdata preprocessing; determining a plurality of Canopy and Canopy center points by adopting a Canopy algorithm based on the minimum and maximum principle; (2) after the step (1), carrying out Kmeans secondary clustering; collecting Canopy center points in the step (1), wherein the number of the Canopy center points is the K value of the Kmeans algorithm; and (3) after the step (2), respectively calculating Euclidean distances from each data point to K clustering center points, and dividing the Euclidean distances into a cluster with the minimum distance to form a new cluster. The road networksubarea division method based on a Canopy-Kmeans clustering algorithm is provided by taking road section center longitude and latitude, road section average speed and road section average density which are acquired in real time as sample data, so that the defects of the Kmeans algorithm are overcome.

Description

technical field [0001] The present invention relates to the field of neural network technology and methods, and more specifically, to a road network subdivision and evaluation method based on Canopy+Kmeans clustering. Background technique [0002] With the increasing size of cities, factors such as uneven distribution of road network congestion, and various types of roads, the urban road network is heterogeneous, which is not conducive to urban traffic management and control. Therefore, it is necessary to subdivide the urban road network Division. The division of road network sub-regions can be regarded as the process of dividing and clustering road segments with similar attributes, so a clustering algorithm can be used to divide road network sub-regions. Clustering is to divide the data set into several clusters (classes) with similar attributes and characteristics. The attributes or characteristics of objects in the same cluster are similar to each other, and are quite di...

AI Technical Summary

Problems solved by technology

Among them, Wang Xiaoxuan (2017) proposed a road network partition method based on Kmeans clustering, but the K value of this method is preset, and the K value is generally difficult to estimate and does not have universality. At the same time, the cluster center is randomly selected, and each time Different clustering results may appear during the operation, and the obtained results may not be the optimal solution

Images