Analysis and Prediction of Spatiotemporal Impact of Traffic Incidents for Better Mobility and Safety inTransportation Systems

The goal of this research is to develop a machine learning framework to predict the spatiotemporal impact of traffic accidents on the upstream traffic and surrounding region. The main objective of the framework is, given a road accident, to forecast when and how the travel-time delay will occur on transportation network. Towards this end, the authors have developed a Dynamic Topology-aware Temporal (DTT) machine learning algorithm that learns the behavior of traffic in both normal conditions and during accidents from the historical traffic sensor datasets. This research exploits four years of real-world Los Angeles traffic sensor data and California Highway Patrol (CHP) accidents logs collected from Regional Integration of Intelligent Transportation Systems (RIITS) under Archived Traffic Data Management System (ADMS) project.

[1]  Jure Leskovec,et al.  Overlapping community detection at scale: a nonnegative matrix factorization approach , 2013, WSDM.

[2]  Christian S. Jensen,et al.  Travel Cost Inference from Sparse, Spatio-Temporally Correlated Time Series Using Markov Models , 2013, Proc. VLDB Endow..

[3]  Shuigeng Zhou,et al.  Shortest Path and Distance Queries on Road Networks: An Experimental Evaluation , 2012, Proc. VLDB Endow..

[4]  Tao Cheng,et al.  Non-parametric regression for space-time forecasting under missing data , 2012, Comput. Environ. Urban Syst..

[5]  Tamara G. Kolda,et al.  Scalable Tensor Factorizations for Incomplete Data , 2010, ArXiv.

[6]  Indre Zliobaite,et al.  Learning under Concept Drift: an Overview , 2010, ArXiv.

[7]  Hyunsoo Kim,et al.  Nonnegative Matrix Factorization Based on Alternating Nonnegativity Constrained Least Squares and Active Set Method , 2008, SIAM J. Matrix Anal. Appl..

[8]  Koby Crammer,et al.  Online Passive-Aggressive Algorithms , 2003, J. Mach. Learn. Res..

[9]  Kristina Lerman,et al.  Tripartite graph clustering for dynamic sentiment analysis on social media , 2014, SIGMOD Conference.

[10]  Anthony K. H. Tung,et al.  SMiLer: A Semi-Lazy Time Series Prediction System for Sensors , 2015, SIGMOD Conference.

[11]  R. K. Agrawal,et al.  A Novel Weighted Ensemble Technique for Time Series Forecasting , 2012, PAKDD.

[12]  Naonori Ueda,et al.  Online Passive-Aggressive Algorithms for Non-Negative Matrix Factorization and Completion , 2014, AISTATS.

[13]  Ee-Peng Lim,et al.  Modeling Temporal Adoptions Using Dynamic Matrix Factorization , 2013, 2013 IEEE 13th International Conference on Data Mining.

[14]  Ugur Demiryurek,et al.  Utilizing Real-World Transportation Data for Accurate Traffic Prediction , 2012, 2012 IEEE 12th International Conference on Data Mining.

[15]  Yi Zhang,et al.  A BPCA based missing value imputing method for traffic flow volume data , 2008, 2008 IEEE Intelligent Vehicles Symposium.

[16]  Yu Zheng,et al.  Travel time estimation of a path using sparse trajectories , 2014, KDD.

[17]  Vikas Sindhwani,et al.  Learning evolving and emerging topics in social media: a dynamic nmf approach with temporal regularization , 2012, WSDM '12.

[18]  Thomas S. Huang,et al.  Graph Regularized Nonnegative Matrix Factorization for Data Representation. , 2011, IEEE transactions on pattern analysis and machine intelligence.

[19]  Lionel M. Ni,et al.  Time-Dependent Trajectory Regression on Road Networks via Multi-Task Learning , 2013, AAAI.

[20]  Dit-Yan Yeung,et al.  Overlapping community detection via bounded nonnegative matrix tri-factorization , 2012, KDD.

[21]  Pang-Ning Tan,et al.  Semi-supervised learning with data calibration for long-term time series forecasting , 2008, KDD.

[22]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[23]  R. Lambiotte,et al.  Random Walks, Markov Processes and the Multiscale Modular Organization of Complex Networks , 2008, IEEE Transactions on Network Science and Engineering.

[24]  Gunnar Rätsch,et al.  Predicting Time Series with Support Vector Machines , 1997, ICANN.

[25]  Chetan Gupta,et al.  Forecasting Spatiotemporal Impact of Traffic Incidents on Road Networks , 2013, 2013 IEEE 13th International Conference on Data Mining.

[26]  Christian S. Jensen,et al.  Using Incomplete Information for Complete Weight Annotation of Road Networks , 2013, IEEE Transactions on Knowledge and Data Engineering.

[27]  James Bailey,et al.  Time Series Forecasting Using Distribution Enhanced Linear Regression , 2013, PAKDD.

[28]  Xiuzhen Cheng,et al.  Traffic clustering and online traffic prediction in vehicle networks: A social influence perspective , 2012, 2012 Proceedings IEEE INFOCOM.

[29]  Masashi Sugiyama,et al.  Trajectory Regression on Road Networks , 2011, AAAI.