基于模糊聚类的不同交通方式的交通事故特征分析

交通事故频发给人们的财产和人身安全带来了极大的危害,掌握交通事故特征对于预防交通事故发生、减少人身伤亡和财产损失至关重要.运用模糊聚类的数量积法对我国2010年交通事故相关数据进行分析,建立模糊相似矩阵,然后用平方法计算传递闭包,得到模糊等价矩阵,根据相似系数不同的值进行动态聚类.结果表明：在五类交通方式中,汽车的事故指标较高,拖拉机、非机动车和行人乘车人的事故指标比较接近,而摩托车的事故指标较低.

Analysis of Characteristics of Traffic Accidents in the Different Modes of Transportation Based on the Fuzzy Clustering

Traffic accidents cause huge property and life losses, so it is important to study the characteristics of the traffic accidents to reduce the losses and prevent the accidents happening. Applying vectors＇ inner product method of fuzzy clustering, the traffic accidents data in 2010 in China is analyzed, and the fuzzy similar matrix is established. Then the square method was applied to calculate transitive closure, and the fuzzy equivalent matrix is obtained. Based on the different values of similarity coefficient, dynamic cluster analysis is carried on. Research results indicate that comparing the traffic accident index of the five kinds of traffic modes, namely automobile, motorcycle, tractor, non-motorized vehicle, and pedestrian, traffic accident probability of automobile is the highest, traffic accident probability of tractor, non-motorized vehicle, and pedestrian is similar and in middle level, and traffic accident probability of motorcycle is the lowest.