A Hybrid Method for Traffic Incident Detection Using Random Forest-Recursive Feature Elimination and Long Short-Term Memory Network With Bayesian Optimization Algorithm

Automatic Incident Detection (AID) is an important part of Intelligent Transportation Systems (ITS). A hybrid AID method using Random Forest-Recursive Feature Elimination (RF-RFE) algorithm and Long-Short Term Memory (LSTM) network optimized by Bayesian Optimization Algorithm (BOA) is proposed in this article. Firstly, a relatively comprehensive set of initial variables is constructed using basic traffic variables and their combinations. Secondly, feature variables are selected from the initial variables using the RF-RFE algorithm. Then, the feature variables are used for training the LSTM network, and the hyper-parameters of the LSTM network are optimized by BOA. In addition, Synthetic Minority Over-Sampling Technique (SMOTE) is employed to solve the problem of imbalance between incident sample size and non-incident sample size. Finally, experiments are conducted using real-world data to test performance of the proposed method and compare with several state-of-the-art AID methods on multiple evaluation criteria. The experimental results illustrate that the proposed method achieved superior performance with respect to almost all the evaluation criteria. It also shows that the proposed method is promising for dealing with the problems of imbalance and small sample size of traffic incident data.

[1]  Yonggang Wang,et al.  A deep fusion model based on restricted Boltzmann machines for traffic accident duration prediction , 2020, Eng. Appl. Artif. Intell..

[2]  Sybil Derrible,et al.  Real-time accident detection: Coping with imbalanced data. , 2019, Accident; analysis and prevention.

[3]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[4]  Junzhou Chen,et al.  Learning Traffic as Images for Incident Detection Using Convolutional Neural Networks , 2020, IEEE Access.

[5]  Bin Ran,et al.  Automated traffic incident detection with a smaller dataset based on generative adversarial networks. , 2020, Accident; analysis and prevention.

[6]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[7]  Bo Peng,et al.  A Deep Cycle Limit Learning Machine Method for Urban Expressway Traffic Incident Detection , 2020 .

[8]  Yanyan Chen,et al.  T-LSTM: A Long Short-Term Memory Neural Network Enhanced by Temporal Information for Traffic Flow Prediction , 2019, IEEE Access.

[9]  Harold J. Paynet.  Freeway Incident-Detection Algorithms Based on Decision Trees With States , 2022 .

[10]  Yuncai Liu,et al.  Traffic Incident Detection Using Multiple-Kernel Support Vector Machine , 2012 .

[11]  Dinesh Singh,et al.  Deep Spatio-Temporal Representation for Detection of Road Accidents Using Stacked Autoencoder , 2019, IEEE Transactions on Intelligent Transportation Systems.

[12]  F. Hutter,et al.  Fast Bayesian hyperparameter optimization on large datasets , 2017, Electronic Journal of Statistics.

[13]  Shuyan Chen,et al.  Multiple Naïve Bayes Classifiers Ensemble for Traffic Incident Detection , 2014 .

[14]  Yasuo Asakura,et al.  Incident Detection Methods Using Probe Vehicles with On-board GPS Equipment , 2015 .

[15]  A. R. Cook,et al.  DETECTION OF FREEWAY CAPACITY-REDUCING INCIDENTS BY TRAFFIC-STREAM MEASUREMENTS , 1974 .

[16]  Shuming Tang,et al.  Traffic-incident detection-algorithm based on nonparametric regression , 2005, IEEE Transactions on Intelligent Transportation Systems.

[17]  Fernando Bação,et al.  Oversampling for Imbalanced Learning Based on K-Means and SMOTE , 2017, Inf. Sci..

[18]  Hussein Dia,et al.  Development and evaluation of arterial incident detection models using fusion of simulated probe vehicle and loop detector data , 2011, Inf. Fusion.

[19]  Xin Jin,et al.  Evaluation of adaptive neural network models for freeway incident detection , 2002, IEEE Transactions on Intelligent Transportation Systems.

[20]  Jianjun Shi,et al.  Detecting and positioning of traffic incidents via video-based analysis of traffic states in a road segment , 2016 .

[21]  Jaouad Boumhidi,et al.  Fuzzy deep learning based urban traffic incident detection , 2017, Cognitive Systems Research.

[22]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[23]  Chinmay Hegde,et al.  Data-driven parallelizable traffic incident detection using spatio-temporally denoised robust thresholds , 2019, Transportation Research Part C: Emerging Technologies.

[24]  Carroll J Messer,et al.  Incident detection on urban freeways , 1974 .

[25]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[26]  Shuo Wang,et al.  Highway crash detection and risk estimation using deep learning. , 2019, Accident; analysis and prevention.

[27]  Fang Yuan,et al.  INCIDENT DETECTION USING SUPPORT VECTOR MACHINES , 2003 .

[28]  Wei Wang,et al.  Decision tree learning for freeway automatic incident detection , 2009, Expert Syst. Appl..

[29]  Baozhen Yao,et al.  A support vector machine with the tabu search algorithm for freeway incident detection , 2014, Int. J. Appl. Math. Comput. Sci..

[30]  Yang Liu,et al.  Multi-class sentiment classification: The experimental comparisons of feature selection and machine learning algorithms , 2017, Expert Syst. Appl..

[31]  Yi Qi,et al.  Application of wavelet technique to freeway incident detection , 2003 .

[32]  Emma E. Regentova,et al.  A Hybrid Model Using Logistic Regression and Wavelet Transformation to Detect Traffic Incidents , 2016 .

[33]  Yorgos J. Stephanedes,et al.  Application of Filtering Techniques for Incident Detection , 1993 .

[34]  Yan Tian,et al.  LSTM-based traffic flow prediction with missing data , 2018, Neurocomputing.

[35]  Jun Cai,et al.  Video-Based Automatic Incident Detection for Smart Roads: The Outdoor Environmental Challenges Regarding False Alarms , 2008, IEEE Transactions on Intelligent Transportation Systems.

[36]  Hojjat Adeli,et al.  Comparison of fuzzy-wavelet radial basis function neural network freeway incident detection model with California algorithm , 2002 .

[37]  Zhongsheng Hua,et al.  A Hybrid Approach for Automatic Incident Detection , 2013, IEEE Transactions on Intelligent Transportation Systems.

[38]  Lee D. Han,et al.  A wavelet-based freeway incident detection algorithm with adapting threshold parameters , 2011 .

[39]  Jianli Xiao,et al.  SVM and KNN ensemble learning for traffic incident detection , 2019, Physica A: Statistical Mechanics and its Applications.

[40]  Fred L. Hall,et al.  Catastrophe theory and patterns in 30 second freeway traffic data, implications for incident detection , 1989 .

[41]  Song Gao,et al.  A Hybrid Method for Traffic Incident Duration Prediction Using BOA-Optimized Random Forest Combined with Neighborhood Components Analysis , 2019, Journal of Advanced Transportation.

[42]  Francesco Marcelloni,et al.  Detection of traffic congestion and incidents from GPS trace analysis , 2017, Expert Syst. Appl..

[43]  Yuncai Liu,et al.  More robust and better: a multiple kernel support vector machine ensemble approach for traffic incident detection , 2014 .

[44]  Soumik Sarkar,et al.  Traffic Congestion Detection from Camera Images using Deep Convolution Neural Networks , 2018, Transportation Research Record: Journal of the Transportation Research Board.

[45]  Wei Wang,et al.  Construct support vector machine ensemble to detect traffic incident , 2009, Expert Syst. Appl..

[46]  John F. Roddick,et al.  An Evaluation of HTM and LSTM for Short-Term Arterial Traffic Flow Prediction , 2019, IEEE Transactions on Intelligent Transportation Systems.

[47]  Henry S. L. Fan,et al.  Development of Dual-Station Automated Expressway Incident Detection Algorithms , 2007, IEEE Transactions on Intelligent Transportation Systems.