In recent years, online ride-hailing services have emerged as an important component of urban transportation system, which not only provide significant ease for residents' travel activities, but also shape new travel behavior and diversify urban mobility patterns. This study provides a thorough review of machine-learning-based methodologies for on-demand ride-hailing services. The importance of on-demand ride-hailing services in the spatio-temporal dynamics of urban traffic is first highlighted, with machine-learning-based macro-level ride-hailing research demonstrating its value in guiding the design, planning, operation, and control of urban intelligent transportation systems. Then, the research on travel behavior from the perspective of individual mobility patterns, including carpooling behavior and modal choice behavior, is summarized. In addition, existing studies on order matching and vehicle dispatching strategies, which are among the most important components of on-line ride-hailing systems, are collected and summarized. Finally, some of the critical challenges and opportunities in ride-hailing services are discussed.

Real-time traffic state (e.g., speed) prediction is an essential component for traffic control and management in an urban road network. How to build an effective large-scale traffic state prediction system is a challenging but highly valuable problem. This study focuses on the construction of an effective solution designed for spatio-temporal data to predict the traffic state of large-scale traffic systems. In this study, we first summarize the three challenges faced by large-scale traffic state prediction, i.e., scale, granularity, and sparsity. Based on the domain knowledge of traffic engineering, the propagation of traffic states along the road network is theoretically analyzed, which are elaborated in aspects of the temporal and spatial propagation of traffic state, traffic state experience replay, and multi-source data fusion. A deep learning architecture, termed as Deep Traffic State Prediction (DeepTSP), is therefore proposed to address the current challenges in traffic state prediction. Experiments demonstrate that the proposed DeepTSP model can effectively predict large-scale traffic states.