EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TAO-ADCNN: Urban traffic congestion prediction and intelligent management in big data environment

Xiang Luo, Shun Liang Chen

Data and Metadata · 08.05.2026 · 5 · 786

DOI: 10.56294/dm2026786 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Data and Metadata, online erschienen am 08.05.2026. Verfasst von Xiang Luo und 1 weiteren Autor:innen. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Traffic congestion is one of the major challenges facing cities experiencing rapid growth due to delay, higher fuel consumption, and harmful environmental impacts. However, large-scale and dynamic traffic data collected from urban environments cannot be effectively analyzed using existing methods, resulting in decreased accuracy, thus requiring an advanced and intelligent traffic forecasting system.

Objective

This research develops a big data–driven urban traffic congestion prediction model using a Termite Alate Optimization-driven Adaptive Deep Convolutional Neural Network (TAO-ADCNN).

Method

Traffic-related data are gathered from publicly available urban traffic repositories, representing diverse road conditions and temporal patterns. During data preprocessing, noise removal, normalization, and missing value treatment are performed using Min–Max normalization and K-Nearest Neighbor enhances [user2.1][a2.2]the quality of the data, and standardizes features for precise prediction. Feature extraction is conducted through Principal Component Analysis (PCA) to capture the most informative traffic flow characteristics while reducing redundancy. In the proposed TAO-ADCNN model, TAO is utilized to adaptively tune network parameters and optimize convolutional layer configurations, improving learning efficiency and convergence, and the entire model is implemented in Python programming environment. The ADCNN then learns complex spatial traffic patterns and temporal congestion indicators from the processed data. This integrated mechanism enables accurate identification of congestion trends across urban road networks.

Results

The proposed method achieves lower MAE (3,8), RMSE (4,5), and a higher F1-score (97%),[user3.1][a3.2] demonstrating superior predictive capability and improved stability compared with conventional approaches.

Conclusion

The modal effectively supports intelligent urban traffic monitoring and proactive congestion management.

Zitieren

APA

Luo, X., & Chen, S. L. (2026). TAO-ADCNN: Urban traffic congestion prediction and intelligent management in big data environment. Data and Metadata, 5, 786. https://doi.org/10.56294/dm2026786

Vancouver

Luo X, Chen SL. TAO-ADCNN: Urban traffic congestion prediction and intelligent management in big data environment. Data and Metadata. 2026;5:786. doi:10.56294/dm2026786

BibTeX
@article{Luo2026,
  author = {Luo, Xiang and Chen, Shun Liang},
  title = {TAO-ADCNN: Urban traffic congestion prediction and intelligent management in big data environment},
  journal = {Data and Metadata},
  year = {2026},
  volume = {5},
  pages = {786},
  doi = {10.56294/dm2026786}
}
Kostenlos starten →