Context-Aware Quantitative Risk Assessment Machine Learning Model for Drivers Distraction
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the critical need for context-aware risk assessment to mitigate accidents caused by driver distraction, particularly as vehicles approach semi-autonomous levels. The authors propose a novel Multi-Class Driver Distraction Risk Assessment (MDDRA) model that categorizes drivers into safe, careless, or dangerous states by integrating vehicle, driver, and environmental data. The methodology utilizes real-world data from the Field Operation Test (TeleFOT) conducted in the East Midlands, UK. The model performs frame-by-frame analysis of driver behavior, incorporating parameters such as hand state, face orientation, eye gaze, road type, weather, and illumination. It calculates a continuous distraction severity score using a weighted average of these contextual factors and aggregates scores over time to determine risk levels. Machine learning techniques are then applied to classify and predict distraction severity to facilitate vehicle takeover when risks are high. The results demonstrate that the Ensemble Bagged Trees algorithm achieved the best performance, with an accuracy of 96.2%. The study confirms that reducing accidents caused by distraction is feasible through this quantitative approach. The model successfully correlates distraction factors with classification severity, providing a robust mechanism for Advanced Driving Assistance Systems (ADAS) to detect risky situations and transition control from the driver to the vehicle.
Key finding
Context-aware risk models significantly outperform static risk assessment approaches by accounting for dynamic environmental factors and individual driver behavior patterns, providing more accurate real-time risk estimates.
Methodology
lab_experiment
Sample size: 45
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via discover_arxiv on 2026-05-04 (4 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | arxiv | — | — | 3 | 2026-05-04 |
| archive | success | — | — | — | 1 | 2026-05-04 |
| extract | success | cached | — | — | 2 | 2026-06-07 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-04 |
| promote | success | — | — | — | 1 | 2026-05-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-07 |
| tag | success | vector_similarity | — | — | 17 | 2026-06-11 |
| verify | success | — | — | — | 1 | 2026-05-08 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-07; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- distraction detection algorithms
- telematics crash prediction
- visual
- situational awareness
- drowsiness detection algorithms
- external distraction
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model, conceptual framework, theory or model