A Data Mining Approach to Identify Key Factors of Traffic Injury Severity

  • Ali Tavakoli Kashani
  • Afshin Shariat-Mohaymany
  • Andishe Ranjbari

Abstract

Seventy percent of the traffic crash fatalities of Iran happen on rural roads, and a significant proportion of the rural roads network of this country is constituted of the main two-lane, two-way roads. The purpose of this study is to identify the most important factors which affect injury severity of drivers involved in traffic crashes on these roads, so that by eliminating or controlling such factors an overall safety improvement can be accomplished. Using the Classification and Regression Tree (CART), one of the powerful data mining tools, the crash data pertaining to the last three years (2006-2008) were analyzed. The variable selection procedure was carried out on the basis of Variable Importance Measure (VIM) which is one of the CART method outputs. The results revealed that not using the seat belt, improper overtaking and speeding are the most important factors associated with injury severity. KEYWORDS: injury severity; traffic safety; data mining; Classification and Regression Trees (CART); Variable Importance Measure (VIM)
Published
2012-01-26
How to Cite
1.
Tavakoli Kashani A, Shariat-Mohaymany A, Ranjbari A. A Data Mining Approach to Identify Key Factors of Traffic Injury Severity. PROMET [Internet]. 2012Jan.26 [cited 2019Dec.9];23(1):11-7. Available from: http://traffic.fpz.hr/index.php/PROMTT/article/view/144
Section
Articles