IBB Traffic Graph Data: Benchmarking and Road Traffic Prediction Model
Eren Olug, Kiymet Kaya, Resul Tugay, Sule Gunduz Oguducu
TL;DR
This paper introduces the IBB Traffic graph dataset for Istanbul to address scale and geographic variability limitations in public road-traffic benchmarks, featuring 2,451 sensors over four years at hourly cadence across urban and highway roads. It presents a Road Traffic Prediction Model that fuses feature engineering, Geometric Laplacian Eigenmap Embedding (GLEE) node representations, and an ExtraTrees classifier to predict congestion as a binary task. Experimental results show that incorporating temporal features and GLEE embeddings yields superior performance, with the best setup achieving an accuracy of about 0.965 and AUROC nearing 0.988, outperforming baselines. The work provides a scalable, geographically diverse benchmark and a modeling approach that can inform intelligent transportation systems and urban traffic management, with future work extending to dynamic graphs.
Abstract
Road traffic congestion prediction is a crucial component of intelligent transportation systems, since it enables proactive traffic management, enhances suburban experience, reduces environmental impact, and improves overall safety and efficiency. Although there are several public datasets, especially for metropolitan areas, these datasets may not be applicable to practical scenarios due to insufficiency in the scale of data (i.e. number of sensors and road links) and several external factors like different characteristics of the target area such as urban, highways and the data collection location. To address this, this paper introduces a novel IBB Traffic graph dataset as an alternative benchmark dataset to mitigate these limitations and enrich the literature with new geographical characteristics. IBB Traffic graph dataset covers the sensor data collected at 2451 distinct locations. Moreover, we propose a novel Road Traffic Prediction Model that strengthens temporal links through feature engineering, node embedding with GLEE to represent inter-related relationships within the traffic network, and traffic prediction with ExtraTrees. The results indicate that the proposed model consistently outperforms the baseline models, demonstrating an average accuracy improvement of 4%.