Presentation of clustering-classification heuristic method for improvement accuracy in classification of severity of road accidents in Iran

• This paper addresses a new clustering-classification method for classifying the severity of road accidents in Iran.
• k-Means and some classification methods are used for running numerical experiments.
• Some activation functions have been used for training the neural networks. The results of them are compared together.
• Also, the accuracy of classification is compared by using different approaches in k-mean.
• Silhouette technique was used to confirm the validity of the clusters.

Data-mining algorithms have been employed in many classification problems. In this essay, a number of these algorithms will be used for classification of road accidents severity (casualties or damages). The individual classification algorithms used in this study including Artificial Neural Network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS). The above-said classification algorithms have some specific advantages and disadvantages in which features and fields of problems influence on their aptness. During recent years, many studies have been carried out on Ensemble Models in order to achieve better results. In this paper, a hybrid idea of clustering-classification method has been adapted by using k-means and Self-Organizing Maps (SOMs) as clustering methods to improve accuracy of classification. The experiments done on the datasets, show that the pre-clustering can improve the accuracy of classification.