Analyzing road accident data using machine learning paradigms

To determine the main factors associated with road traffic accidents is one of main objectives of accident data analysis. Due heterogeneity nature of road accident data makes analysis tricky. To overcome heterogeneity of data partitioning is used. The proposed method uses k-means clustering method as the main task of segmentation of road accident data. Further, association rule mining is applied to discover the situations related with the occurrence of the whole data set and the occurrence of clusters recognized by the k-means clustering algorithm. The combined result of k-means clustering and association rule produces major information.