Title: Traffic accident prediction based on an artificial bee colony algorithm and a self-adaptive fuzzy wavelet neural network

Authors: Zhicheng Li

Addresses: Department of Urban Rail Transit and Information Engineering, Anhui Communications Vocational and Technical College, Hefei, 230051, China

Abstract: This study combines an artificial bee colony algorithm with a fuzzy wavelet neural network to establish a new traffic accident prediction model. It introduces the self-adaptive mutation operation of differential evolution algorithms and the selection operator, crossover operator, and mutation operator of a genetic algorithm into the traditional artificial bee colony algorithm. These actions aim to improve the shortcomings of the slow convergence rate and the weak local search ability of the artificial bee colony algorithm in the later period. This study also uses the improved artificial bee colony algorithm to optimise the weights and thresholds of the fuzzy wavelet neural network and to make up for the randomness of the weights and threshold selection of the wavelet neural network. It combines the good nonlinear fitting ability of the self-adaptive fuzzy wavelet neural network with the strong robustness of the artificial bee colony algorithm to build a model for predicting traffic death tolls. Computer simulation shows that the prediction accuracy of the proposed method is higher.

Keywords: traffic accidents; prediction; improved artificial bee colony algorithm; fuzzy wavelet neural network; differential evolution algorithm.

DOI: 10.1504/IJCSM.2023.131464

International Journal of Computing Science and Mathematics, 2023 Vol.17 No.3, pp.254 - 265

Received: 02 Mar 2022
Accepted: 25 Jul 2022

Published online: 13 Jun 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article