Driving style classification and trip scoring based on Machine learning
Driving styles reflect the behaviours of drivers and their tendencies to drive carefully, perform riskier manoeuvres or break traffic rules. They also play a significant role in traffic safety, as how calm or aggressive a person drives can determine how likely they are to enter an unsafe scenario and potentially crash. There are a wide variety of potential applications for a system that can identify driving styles and score driving trips, such as increasing drivers’ awareness of their shortcomings and adaptive insurance policies. To tackle this topic, this paper details a trip-scoring system using a Siamese network that inputs data grouped using k-means clustering. This model can be used to measure the aggressiveness of drivers in their respective communities by comparing them to their peers in similar environments; Siamese network is a high-speed machine learning system for comparing data that will allow the inclusion of new users into the scoring system without requiring the model to be retrained every time a new user is added.
Authors
- Sébastien Glaser (ORCID: https://orcid.org/0000-0003-0658-7765)
- Mahmoud Masoud (ORCID: https://orcid.org/0000-0002-0130-4327)
- Mohammed Mamdouh Zakaria Elhenawy (ORCID: https://orcid.org/0000-0003-2634-4576)
- Waqar Ahmed Khan (ORCID: https://orcid.org/0000-0002-4082-0893)
- Long Wang (ORCID: https://orcid.org/0000-0001-6695-6054)
Institutions
- King Fahd University of Petroleum and Minerals (SA)
- Queensland University of Technology (AU)
- University of Sharjah (AE)
Publication Details
- Journal
- Transportation Research Interdisciplinary Perspectives
- Published
- 2026-11-13
- DOI
- https://doi.org/10.1016/j.trip.2026.102327
- Primary Topic
- Traffic and Road Safety
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Queensland University of Technology
- King Fahd University of Petroleum and Minerals