ExtTra: Short-Term Traffic Flow Prediction Based on Extremely Randomized Trees
Short-term traffic flow prediction is an important task for intelligent transportation systems. Conventional time series based approaches such as ARIMA can hardly reflect the inter-dependence of related roads. Other parametric or nonparametric methods do not take full advantage of the spatial temporal features. Moreover, some machine learning models are still not investigated in solving this problem. To fill this gap, in this paper we propose ExtTra: an extremely randomized trees based approach for short-term traffic flow prediction. To the best of our knowledge, our work is the first effort to apply the extremely randomized trees model on the traffic flow prediction problem. Moreover, our approach incorporates new spatial temporal features which were not considered in previous studies. Experimental results show that our approach significantly outperforms the baselines in prediction accuracy.
KeywordsTraffic flow prediction Extremely randomized trees Spatial temporal features Intelligent transportation Machine learning
This work was supported in part by: National Natural Science Foundation of China (No. 61702059), Frontier and Application Foundation Research Program of Chongqing City (No. cstc2018jcyjAX0340), Chongqing Industrial Generic Technology Innovation Program (No. cstc2017zdcy-zdzxX0010), Guangxi Key Laboratory of Trusted Software (No. kx201702).
- 4.Pascale, A., Nicoli, M.: Adaptive Bayesian network for traffic flow prediction. In: 2011 IEEE Statistical Signal Processing Workshop (SSP), pp. 177–180 (2011)Google Scholar
- 12.Du, S., Li, T., Gong, X., Yang, Y., Horng, S.J.: Traffic flow forecasting based on hybrid deep learning framework. In: 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE), pp. 1–6 (2017)Google Scholar
- 13.Zhang, J., Zheng, Y., Qi, D.: Deep spatio-temporal residual networks for citywide crowd flows prediction. In: AAAI, pp. 1655–1661 (2017)Google Scholar
- 14.Wang, D., Xiong, J., Xiao, Z., Li, X.: Short-term traffic flow prediction based on ensemble real-time sequential extreme learning machine under non-stationary condition. In: IEEE 83rd Vehicular Technology Conference (VTC Spring), pp. 1–5 (2016)Google Scholar