Multi-modal transportation recommendation with unified route representation learning

Abstract

Multi-modal transportation recommendation aims to provide the most appropriate travel route with various transportation modes according to certain criteria. After analyzing large-scale navigation data, we find that route representations exhibit two patterns: spatio-temporal autocorrelations within transportation networks and the semantic coherence of route sequences. However, there are few studies that consider both patterns when developing multi-modal transportation systems. To this end, in this paper, we study multi-modal transportation recommendation with unified route representation learning by exploiting both spatio-temporal dependencies in transportation networks and the semantic coherence of historical routes. Specifically, we propose to unify both dynamic graph representation learning and hierarchical multi-task learning for multi-modal transportation recommendations. Along this line, we first transform the multi-modal transportation network into time-dependent multi-view transportation graphs and propose a spatiotemporal graph neural network module to capture the spatial and temporal autocorrelation. Then, we introduce a coherent-aware attentive route representation learning module to project arbitrary-length routes into fixed-length representation vectors, with explicit modeling of route coherence from historical routes. Moreover, we develop a hierarchical multi-task learning module to differentiate route representations for different transport modes, and this is guided by the final recommendation feedback as well as multiple auxiliary tasks equipped in different network layers. Extensive experimental results on two large-scale real-world datasets demonstrate the performance of the proposed system outperforms eight baselines.

Times series forecasting for urban building energy consumption based on graph convolutional network

“…In a graph network, GNNs adopt the message passing mechanism, which means that the representation of each node is naturally defined by its attributes and aggregation of the adjacent nodes [41]. Due to the nature of graph network, GNN has been broadly employed in structural scenarios where data structures have explicit relationships, such as social network [42], transportation system [43], and recommendation system algorithm [44,45]. The information propagation allows GNN models to learn building energy consumption based on its own characteristics and propagate the energy impact from neighborhood buildings via graph structure.…”

Weakly-supervised Temporal Path Representation Learning with Contrastive Curriculum Learning -- Extended Version

“…d i represents the rank difference on the i-th competitive path in both rankings. c) Path Recommendation: A similar strategy is used in an existing study [48], where a path is associated with a binary label with the help of users' trajectories. A path used by a user's trajectory, say path A, is labeled 1, whereas alternative paths connecting the same source and destination, say paths B and C, are labeled 0.…”

Intelligent Electric Vehicle Charging Recommendation Based on Multi-Agent Reinforcement Learning

“…We develop a simulator 1 based on historical real-time availability (supplies) records data, historical real-time charging prices data, and charging powers data of charging stations, along with historical electric vehicles (EVs) charging data, and road network data [21], to simulate how the system runs for a day. We take Baidu API 2 to compute ETA [23] between the location of charging request and charging station.…”