Spatial-Temporal Multi-Granularity Framework for Accurate and Robust Traffic Forecasting
A novel multi-granularity framework that enhances the capture of long-distance and long-term dependencies in traffic networks, enabling accurate and robust traffic forecasting.
Short-term Prediction of Construction Waste Transport Activities Using an AI-Powered Ensemble Learning Framework
An AI-powered ensemble learning framework, coined AI-Truck, is designed to accurately predict the levels of construction waste transport activities at a city scale during heavy pollution episodes, enabling timely and proactive environmental law enforcement.
Explainable and Accurate Traffic Flow Prediction with Large Language Models
Large language models can effectively capture complex spatio-temporal patterns in traffic data and generate interpretable predictions, outperforming state-of-the-art deep learning models.
Semantic-Fused Multi-Granularity Cross-City Traffic Prediction: Enhancing Urban Mobility through Knowledge Transfer
The core message of this paper is to propose a Semantic-Fused Multi-Granularity Transfer Learning (SFMGTL) model that can effectively leverage knowledge from data-rich source cities to improve traffic demand prediction in data-scarce target cities. The model dynamically fuses multiple urban semantics, learns hierarchical node clustering, and extracts domain-invariant meta-knowledge to enable robust cross-city knowledge transfer.
Unifying Lane-Level Traffic Prediction: Benchmark and Baseline from a Graph Structural Perspective
Lane-level traffic prediction is crucial for intelligent transportation systems, and this paper introduces a unified spatial topology structure and a simple baseline model, GraphMLP, to enhance lane-level traffic prediction accuracy.
BjTT: A Large-scale Multimodal Dataset for Traffic Prediction
Urban transportation systems face challenges in accurate traffic prediction due to various factors. The BjTT dataset introduces a large-scale multimodal approach to enhance traffic prediction accuracy.
Spatio-Temporal Self-Supervised Learning for Traffic Flow Prediction: Enhancing Spatial and Temporal Heterogeneity Modeling
The author proposes a novel Spatio-Temporal Self-Supervised Learning (ST-SSL) framework to enhance traffic pattern representations by addressing spatial and temporal heterogeneity through adaptive self-supervised learning paradigms.
© 2024 by Linnk AI