A study on road accident prediction and contributing factors using explainable machine learning models: analysis and performance

Ahmed, Shakil; Hossain, Akbar; Ray, Sayan Kumar; Bhuiyan, Md. Mafijul Islam; Sabuj, Saifur Rahman

doi:10.1016/j.trip.2023.100814

Cited by 25 publications

(17 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Men, especially those under 25, are disproportionately involved in RTAs. They are responsible for 73% of deaths and have a threefold higher risk than their female counterparts [5] . Various reasons contribute to this gender gap, including mood stability, driving habits, and alcohol and tobacco use.…”

Section: Policy Implicationsmentioning

confidence: 99%

Guardians of the road: machine learning solutions for safer commutes

PENG,

He,

Gao

et al. 2024

Fourth International Conference on Smart City Engineering and Public Transportation (SCEPT 2024)

View full text Add to dashboard Cite

Section: Policy Implicationsmentioning

confidence: 99%

Guardians of the road: machine learning solutions for safer commutes

PENG,

He,

Gao

et al. 2024

Fourth International Conference on Smart City Engineering and Public Transportation (SCEPT 2024)

View full text Add to dashboard Cite

“…Machine Learning (ML) models to predict road accident severity using New Zealand's Road accident data from 2016-2020 (12) . The models used include Random Forest (RF), Decision Jungle (DJ), Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (L-GBM), and Categorical Boosting (CatBoost).…”

Section: Background Workmentioning

confidence: 99%

“…The key insight from the Tables 4, 5 and 6 and Figure 2 shows that the Random Forest is the most effective algorithm among the three ML algorithms for performing this particular task. (12) ML models (RF, DJ, AdaBoost, XGBoost, L-GBM, CatBoost)…”

Section: Fig 2 Accuracy Of Various ML Algorithmsmentioning

confidence: 99%

“…Table 7 provides a concise summary of each study, the methods used, their key findings, and the accuracy of their models. The studies by Shakil et al (12) , Josephinshermila et al (2) and Krioudj et al (10) are particularly relevant as they also use machine learning models and provide accuracy metrics, which can be directly compared with the accuracy of the proposed model, and https://www.indjst.org/ it performs better than previous accident detection system. This might entail examining research that has employed identical or comparable machine learning algorithms for detecting accidents.…”

Section: 02% (Rf)mentioning

confidence: 99%

See 1 more Smart Citation

Instinctive Detection of Accident Occurrence using Numerous Machine Learning Techniques with Comparative Study

Rajamanogaran,

Karthikeyan

2024

IJST

View full text Add to dashboard Cite

Objectives: The research work aims to develop an automated detection system that uses video captures to identify roadside accidents or significant events. By alerting nearby hospitals and emergency services, the system reduces response times and potentially saves lives. The system’s integration with existing emergency response systems ensures prompt assistance to those in need. Methods: The video processing pipeline begins by converting video files into frames for analysis and feature extraction. These features serve as inputs for classification algorithms such as Random Forest, SVM, and KNN. The model’s performance is evaluated using a training set and unseen test data, with the predicted classifications compared against the ground truth labels. Findings: Among the tested classification algorithms, the Random Forest algorithm achieved the highest accuracy. Using 128 frames for analysis provided more comprehensive information, yielding a 96% accuracy rate. This combination proves to be a powerful tool in classification tasks, providing reliable and accurate outputs. Novelty: Machine learning algorithms are instrumental in automating accident detection from video captures. They analyse video footage to identify accidents and promptly alert relevant authorities. This technology can also dispatch emergency messages to nearby hospitals, ensuring quick assistance. The consideration of different frame counts in classification improves accuracy by capturing critical moments and patterns. Machine learning algorithms applied in this work significantly enhance emergency response, reduce response times, and potentially save lives. Keywords: Emergency Services, Accidents, Feature Extraction, Random Forest, SVM, KNN

show abstract

“…However, studies comparing the traditional models with ML models report contrary result. While some studies report ML algorithms of Random Forest reporting higher accuracy [ 30 ], other studies report that the traditional models report the best performance [ 31 ]. While our current study doesn't involve a comparison with traditional models, it focuses on evaluating various ML models to identify models that demonstrates superior performance.…”

Section: Introductionmentioning

confidence: 99%

Severity of vehicle-to-vehicle accidents in the UAE: An exploratory analysis using machine learning algorithms

Maghelal,

Ali,

Azar

et al. 2023

Heliyon

View full text Add to dashboard Cite

A study on road accident prediction and contributing factors using explainable machine learning models: analysis and performance

Cited by 25 publications

References 28 publications

Guardians of the road: machine learning solutions for safer commutes

Guardians of the road: machine learning solutions for safer commutes

Instinctive Detection of Accident Occurrence using Numerous Machine Learning Techniques with Comparative Study

Severity of vehicle-to-vehicle accidents in the UAE: An exploratory analysis using machine learning algorithms

Contact Info

Product

Resources

About