基于灰色关联与Apriori算法的道路交通事故数据分析

Analysis of Road Traffic Accident Data Based on Grey Relational Analysis and Apriori Algorith

针对运用关联规则挖掘道路交通事故(road traffic accident,简称RTA)数据产生大量无效规则与使用灰色关联分析无法发现数据隐藏联系的缺陷,提出将灰色关联分析与关联规则的Apriori算法相结合的方法,并将其运用于RTA数据。首先进行数据预处理;然后运用灰色关联分析,找到与RTA发生数有强关联的因素,并保留其数据,反之剔除相关数据;最后对保留下来的数据进行关联规则挖掘,找到数据之间令人感兴趣的联系。将此方法运用在四川省RTA数据上,通过与仅采用Apriori算法相对比,实验结果表明:此算法的时间缩短了近40%,冗余规则减少了近50%。由此挖掘出了RTA数据之间有趣的联系,并证明了灰色关联分析与Apriori算法相结合方法的有效性。

In view of the application of association rules to mining road traffic accident(referred to as RTA)data,a large number of invalid rules are produced and the data hiding links can not be found by using grey correlation analysis.The method of combining grey relational analysis with Apriori algorithm of association rules is proposed and applied to RTA data.First,the data preprocessing is carried out;then the grey relational analysis is used to find the factors which are strongly associated with the RTA occurrence,and retain their data,and otherwise eliminate the relevant data;finally,the association rules are excavated for the retained data to find an interesting connection between the data.This method is applied to the RTA data of Sichuan province.By comparing with the only Apriori algorithm,the experimental results show that the time of the algorithm is shortened by nearly 40%,and the redundancy rule is reduced by nearly 50%.From this,we find out interesting links between RTA data,and prove the validity of the combination method of grey relational analysis and association rules.