Shuqi Zhong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shuqi Zhong has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Jiatao Chen
School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China
1 shared publication - Ming Cai
School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen, China
1 shared publication
Latest Publications
- Open Access
- 5 December 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (23) , 3843
A Transport Mode Detection Framework Based on Mobile Phone Signaling Data Combined with Bus GPS Data
Transport mode is one of the important travel characteristics for citizens, which is crucial to the planning and management of urban transportation. With the natural advantages of large sample sizes and a wide coverage of people, more and more researchers adopt mobile phone signaling data (MSD) to detect transport modes. However, due to their low positioning accuracy and temporally irregular nature, identifying transport modes with similar spatiotemporal features, such as the bus and car modes, is particularly challenging. We propose a transport detection framework using MSD combined with bus GPS data to distinguish between the car and bus modes. First, a trajectory matching algorithm is proposed to obtain the most probable bus that mobile phone users may take. Then, more features are mined to improve the accuracy of transport mode detection with different classification models. Furthermore, for signaling trajectories identified as the bus mode, more bus travel information is recognized, including the boarding and alighting station and timestamp. Finally, we built a ground truth dataset and compared the recognition accuracies under different features and classification models. The result shows that the transport mode detection accuracies of the proposed framework with the GBDT, XGBoost, and LightGBM algorithms are all higher than 94%.
- 5 December 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (23) , 3843
Open AccessA Transport Mode Detection Framework Based on Mobile Phone Signaling Data Combined with Bus GPS Data
Transport mode is one of the important travel characteristics for citizens, which is crucial to the planning and management of urban transportation. With the natural advantages of large sample sizes and a wide coverage of people, more and more researchers adopt mobile phone signaling data (MSD) to detect transport modes. However, due to their low positioning accuracy and temporally irregular nature, identifying transport modes with similar spatiotemporal features, such as the bus and car modes, is particularly challenging. We propose a transport detection framework using MSD combined with bus GPS data to distinguish between the car and bus modes. First, a trajectory matching algorithm is proposed to obtain the most probable bus that mobile phone users may take. Then, more features are mined to improve the accuracy of transport mode detection with different classification models. Furthermore, for signaling trajectories identified as the bus mode, more bus travel information is recognized, including the boarding and alighting station and timestamp. Finally, we built a ground truth dataset and compared the recognition accuracies under different features and classification models. The result shows that the transport mode detection accuracies of the proposed framework with the GBDT, XGBoost, and LightGBM algorithms are all higher than 94%.