According to World Health Organization, each year, over 1.2 million people die on roads, and between 20 and 50 million suffer non-fatal injuries. Based on international reports, Iran has a high death rate caused by road accidents. The objective of this study was to extract implicit knowledge from road accident data sets on roads of Iran through data mining. In this regard, three useful data mining techniques were combined: clustering, classification and rule extraction. Following the preparation stage, data were segmented via three clustering algorithms; Kohonen, K-Means and Twostep. Two-step cluster analysis is a one-pass-through data approach which generates a fairly large number of pre-clusters. Next, the optimized algorithm and cluster were identified, after which, in the classification level and by adding the drivers' demographic features through C5.0, a classification algorithm was employed so as to make the decision tree. Ultimately, the effects of these demographic features were investigated on road accidents. The characteristics such as age, job, driving license duration and gender proved to be more important factors in accident analysis. Certain rules of accidents were then extracted in each season of the year.
The Effect of Drivers' Demographic Characteristics on Road Accidents in Different Seasons Using Data Mining
2017
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Papers - Women drivers' behaviour, socio-demographic characteristics and accidents
Online Contents | 1999
|Study on Road Accidents Using Data Mining Technology
IEEE | 2018
|Analysis of Road Accidents Using Data Mining Paradigm
Springer Verlag | 2021
|Papers - Road accidents caused by drivers falling asleep
Online Contents | 1999
|