A Cloud-Based Environment-Aware Driver Profiling Framework using Ensemble Supervised Learning

Driver profiling is an emerging scheme that has a wide range of applications in the field of Intelligent Transportation Systems (ITS). Driver profiling is the real-time process of detecting driving behaviors and computing a driver's competence level based on detected behaviors. In this paper, a novel driver profiling framework is presented. A risk prediction model is hosted in the cloud to determine the risk associated with detected behaviors in specific driving environments. Risk values along with a driver's compliance to warnings are both utilized to compute a driver's risk profile. Using SHRP2 large-scale Naturalistic Driving (ND) dataset, the development of the risk prediction model is presented herein with the underlying sub-processes of data preprocessing, error analysis, and model selection. Validation results show that a developed randomized trees supervised learning model is proven to have a good tradeoff between bias and variance with evidently high performance results.