Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)

<Previous Article In Volume
Next Article In Volume>

A Review of Recent Methods for Traffic Prediction with Few Data

Authors
Zhaowei Huang1, *
1Mathematics Department, Santa Monica College, Santa Monica, CA, USA
*Corresponding author. Email: huang_zhaowei01@student.smc.edu
Corresponding Author
Zhaowei Huang
Available Online 24 April 2026.
DOI
10.2991/978-94-6239-648-7_80How to use a DOI?
Keywords
Traffic Prediction; Few-Shot Learning; Zero-Shot Learning; Meta-Learning; Graph Neural Networks
Abstract

Predicting traffic accurately is very important for building smart cities. But the deep learning models that achieve state-of-the-art performance today—especially graph neural networks (GNNs)—require massive amounts of past data. This is a problem in new urban areas or on roads that have recently been built—where the data doesn’t yet exist. It’s for this reason that Few-Shot Learning (FSL) and Zero-Shot Learning (ZSL) have played such a crucial role in recent traffic prediction research. In this survey, we examine four recent strategies that are practical and promising: (1) meta-learning, which helps models adapt quickly to new situations; (2) better GNNs, which we build by designing them with how traffic actually behaves in mind; (3) large foundation models trained on transportation data, and (4) the surprising use of large language models (LLMs) for forecasting—how each works, what they’re good at, and where they fail. Our review makes clear where we’re at and where we’re going, and is intended for anyone who wants to build traffic predictors that work, even when data is hard to come by.

Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
Series
Advances in Computer Science Research
Publication Date
24 April 2026
ISBN
978-94-6239-648-7
ISSN
2352-538X
DOI
10.2991/978-94-6239-648-7_80How to use a DOI?
Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - Zhaowei Huang
PY  - 2026
DA  - 2026/04/24
TI  - A Review of Recent Methods for Traffic Prediction with Few Data
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 739
EP  - 744
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_80
DO  - 10.2991/978-94-6239-648-7_80
ID  - Huang2026
ER  -