دانلود رایگان مقاله انگلیسی مدلسازی فرکانس تصادف توسط شبکه عصبی بهینه - نشریه الزویر

Abstract

This study develops a neural network (NN) model to explore the nonlinear relationship between crash frequency and risk factors. To eliminate the possibility of over-fitting and to deal with the black-box characteristic, a network structure optimization algorithm and a rule extraction method are proposed. A case study compares the performance of the trained and modified NN models with that of the traditional negative binomial (NB) model for analyzing crash frequency on road segments in Hong Kong. The results indicate that the optimized NNs have somewhat better fitting and predictive performance than the NB models. Moreover, the smaller training/testing errors in the optimized NNs with pruned input and hidden nodes demonstrate the ability of the structure optimization algorithm to identify the insignificant factors and to improve the model generalization capacity. Furthermore, the rule-set extracted from the optimized NN model can reveal the effect of each explanatory variable on the crash frequency under different conditions, and implies the existence of nonlinear relationship between factors and crash frequency. With the structure optimization algorithm and rule extraction method, the modified NN model has great potential for modeling crash frequency, and may be considered as a good alternative for road safety analysis.

6. Conclusions and future research

This study develops an NN for modeling possible nonlinear relationship between crash frequency and risk factors. To improve the generalization capacity and to deal with the black-box characteristic of the NN, a structure optimization N2PFA algorithm and a modified rule extraction algorithm are proposed. A crash dataset obtained from the TIS maintained by the Transport Department of Hong Kong is used to demonstrate the proposed methods and to compare them with the results of an NB model. The results show that both the trained and optimized NNs outperform the NB models in fitting and predictive performance to somewhat extent. In the optimized NNs, certain numbers of input and hidden nodes are dropped off, and better approximation performance is achieved, demonstrating the ability of the N2PFA algorithm to identify insignificantfactors and to improve the model generalization capacity. The optimized NN generates ten rules in which the coefficients of the explanatory variables are different, which confirms that they are nonlinearly related to the crash frequency. The signs of these coefficients have identical directions under most conditions, and are consistent with those in the NB model. Moreover, most of the results for the explanatory variables are reasonable and conform to traffic engineering experience or the findings of previous studies, which further validates the proposed method