EUVIMEDEUVIMED entdecken
Crossref

Urban traffic flow prediction and traffic state identification using catboost with shap based analysis

Ly Pham Thi, Du Nguyen Duc, Hoa Nguyen Thi Hong

Transport and Communications Science Journal · 15.03.2026 · 77 (4) · 571-582

DOI: 10.47869/tcsj.77.4.17 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transport and Communications Science Journal, online erschienen am 15.03.2026. Verfasst von Ly Pham Thi und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traffic flow forecasting is a critical component in intelligent transportation systems, supporting traffic management, reducing congestion, and improving the operational efficiency of urban road networks. However, this is a challenging problem due to the temporal variability of traffic data and the influence of numerous complex factors. In this study, we propose a traffic flow forecasting method based on the CatBoost algorithm to effectively exploit tabular traffic data collected from traffic sensors. The dataset consists of 2,976 records containing temporal information and vehicle counts across four categories (cars, motorcycles, buses, and trucks). In addition to the original features, the study constructs supplementary temporal and time-series features, including Hour, DayOfWeek, IsWeekend, Total_lag1, and Total_roll3, to enhance the model's ability to capture traffic flow variation trends. Based on this, two independent machine learning tasks are established: (i) total traffic flow forecasting as a regression problem, and (ii) traffic condition classification into four levels. Experimental results demonstrate that the proposed model achieves strong predictive performance. Furthermore, feature importance analysis using the SHAP method reveals that vehicle count-related variables, particularly CarCount and BusCount, have a significant impact on prediction outcomes. The study demonstrates that CatBoost is an effective approach for traffic flow forecasting with tabular data and holds strong potential for application in intelligent traffic management systems.

Zitieren

APA

Thi, L. P., Duc, D. N., & Hong, H. N. T. (2026). Urban traffic flow prediction and traffic state identification using catboost with shap based analysis. Transport and Communications Science Journal, 77(4), 571-582. https://doi.org/10.47869/tcsj.77.4.17

Vancouver

Thi LP, Duc DN, Hong HNT. Urban traffic flow prediction and traffic state identification using catboost with shap based analysis. Transport and Communications Science Journal. 2026;77(4):571-582. doi:10.47869/tcsj.77.4.17

BibTeX
@article{Thi2026,
  author = {Thi, Ly Pham and Duc, Du Nguyen and Hong, Hoa Nguyen Thi},
  title = {Urban traffic flow prediction and traffic state identification using catboost with shap based analysis},
  journal = {Transport and Communications Science Journal},
  year = {2026},
  volume = {77},
  number = {4},
  pages = {571--582},
  doi = {10.47869/tcsj.77.4.17}
}
Kostenlos starten →