Analysis and Monitoring of the Traffic Suburban Road Accidents Using Data Mining Techniques; A Case Study of Isfahan Province in Iran

Driving accidents have been always counted as one of the most ostensible causes of deaths in the societies today. Statistics and reports indicate that the road accidents in Iran rank several times more than the ones in the developed countries. In the current paper, the rules and factors influencing the traffic road accidents of Iran have been extracted along with extracting a local data model after collecting the data from a variety of sources followed by data aggregation and combination, data cleaning, and separating the inappropriate data. This was done by employing appropriate data mining methods, such as clustering and decision tree. The utilized data was based on 10000 accidents during 2011 to 2013 in Isfahan Province, Iran. The experimental results have revealed that of the Decision Tree approaches, C5.0 algorithm outperforms the other algorithms with a lower error rate and a higher accuracy rate. Our research analysis also shows that in determining the accident type, three most important attributes include the type of the faulty vehicle, type of the vehicle hit, and the accident reason. The results and findings obtained in this study are significant and interesting which can provide the authorities with invaluable information on reducing the road accidents.