Highway road accident analysis based on clustering ensemble

Abstract

We employ clustering ensemble to partition highway roads according to traffic accident information to avoid the occurrence of accidents in this paper. Above all, we use fuzzy k-means clustering to classify numerical data of accidents for producing numerical clustering membership, and produce categorical memberships using values of corresponding categorical attributes. Then we adopt clustering ensemble to merge all clustering memberships to solve the sole clustering. Finally, the clustering ensemble was used to group 16 highway roads and results show that it is effective and could be used to avoid occurrence of traffic accidents. © 2011 Springer-Verlag Berlin Heidelberg.