Road Crash Injury Severity Prediction Using a Graph Neural Network Framework
Crash severity prediction is a challenging research area, where the objective is to accurately assess the extent of severity of an injury resulting from road traffic accidents. The main aim of existing studies is to precisely assess the potential severity of crashes under diverse circumstances, such...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10460524/ |
| _version_ | 1797243390356619264 |
|---|---|
| author | Karim A. Sattar Iskandar Ishak Lilly Suriani Affendey Siti Nurulain Binti Mohd Rum |
| author_facet | Karim A. Sattar Iskandar Ishak Lilly Suriani Affendey Siti Nurulain Binti Mohd Rum |
| author_sort | Karim A. Sattar |
| collection | DOAJ |
| description | Crash severity prediction is a challenging research area, where the objective is to accurately assess the extent of severity of an injury resulting from road traffic accidents. The main aim of existing studies is to precisely assess the potential severity of crashes under diverse circumstances, such as weather conditions, vehicle attributes, road characteristics and layout, and traffic control factors. This effort aids authorities in establishing effective emergency response systems. The novelty and objective of our work involve contributing to this research area by employing a graph architecture to capture relationships among various crash records to uncover any hidden patterns that traditional ML models might overlook. The current study extends existing knowledge by leveraging Graph Neural Networks (GNN) and comparing their performance to popular ensemble-based models, which include Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Artificial Neural Networks (ANNs). Real data from the United Kingdom (UK) was employed to achieve our goal. The data was obtained from the Department for Transport open data portal. All models underwent training using the training dataset, followed by performance evaluation using diverse metrics such as the accuracy, precision, recall, f1-score, Matthews Correlation Coefficient (MCC), confusion matrix, and computational cost on the test dataset. Overall, our proposed GNN-based model demonstrated better performance when compared to other models. Specifically, the GNN model outperformed all other models across all metrics. For instance, the accuracy of the GNN model was 85.55% as compared to 83.36%, 83.18%, and 83.27% for the XGBoost, RF, and ANN models, respectively. The GNN model assisted in identifying hidden patterns by considering non-linear relationships among crash records. Thus, the model had the potential to improve its ability to predict severe accidents, which could in turn significantly improve emergency response efforts and reduce the likelihood of severe accidents resulting in fatalities. |
| first_indexed | 2024-04-24T18:54:21Z |
| format | Article |
| id | doaj.art-c169ce602ea247788f703a758aabd55c |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-24T18:54:21Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-c169ce602ea247788f703a758aabd55c2024-03-26T17:45:20ZengIEEEIEEE Access2169-35362024-01-0112375403755610.1109/ACCESS.2024.337388510460524Road Crash Injury Severity Prediction Using a Graph Neural Network FrameworkKarim A. Sattar0https://orcid.org/0000-0002-4170-5360Iskandar Ishak1https://orcid.org/0000-0001-8874-1417Lilly Suriani Affendey2https://orcid.org/0000-0001-7947-8792Siti Nurulain Binti Mohd Rum3https://orcid.org/0000-0001-7818-8208Department of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, MalaysiaDepartment of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, MalaysiaDepartment of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, MalaysiaDepartment of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, MalaysiaCrash severity prediction is a challenging research area, where the objective is to accurately assess the extent of severity of an injury resulting from road traffic accidents. The main aim of existing studies is to precisely assess the potential severity of crashes under diverse circumstances, such as weather conditions, vehicle attributes, road characteristics and layout, and traffic control factors. This effort aids authorities in establishing effective emergency response systems. The novelty and objective of our work involve contributing to this research area by employing a graph architecture to capture relationships among various crash records to uncover any hidden patterns that traditional ML models might overlook. The current study extends existing knowledge by leveraging Graph Neural Networks (GNN) and comparing their performance to popular ensemble-based models, which include Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Artificial Neural Networks (ANNs). Real data from the United Kingdom (UK) was employed to achieve our goal. The data was obtained from the Department for Transport open data portal. All models underwent training using the training dataset, followed by performance evaluation using diverse metrics such as the accuracy, precision, recall, f1-score, Matthews Correlation Coefficient (MCC), confusion matrix, and computational cost on the test dataset. Overall, our proposed GNN-based model demonstrated better performance when compared to other models. Specifically, the GNN model outperformed all other models across all metrics. For instance, the accuracy of the GNN model was 85.55% as compared to 83.36%, 83.18%, and 83.27% for the XGBoost, RF, and ANN models, respectively. The GNN model assisted in identifying hidden patterns by considering non-linear relationships among crash records. Thus, the model had the potential to improve its ability to predict severe accidents, which could in turn significantly improve emergency response efforts and reduce the likelihood of severe accidents resulting in fatalities.https://ieeexplore.ieee.org/document/10460524/Categorical embeddinggraph neural networkGraphSAGEkNN graphroad crash injury severity |
| spellingShingle | Karim A. Sattar Iskandar Ishak Lilly Suriani Affendey Siti Nurulain Binti Mohd Rum Road Crash Injury Severity Prediction Using a Graph Neural Network Framework IEEE Access Categorical embedding graph neural network GraphSAGE kNN graph road crash injury severity |
| title | Road Crash Injury Severity Prediction Using a Graph Neural Network Framework |
| title_full | Road Crash Injury Severity Prediction Using a Graph Neural Network Framework |
| title_fullStr | Road Crash Injury Severity Prediction Using a Graph Neural Network Framework |
| title_full_unstemmed | Road Crash Injury Severity Prediction Using a Graph Neural Network Framework |
| title_short | Road Crash Injury Severity Prediction Using a Graph Neural Network Framework |
| title_sort | road crash injury severity prediction using a graph neural network framework |
| topic | Categorical embedding graph neural network GraphSAGE kNN graph road crash injury severity |
| url | https://ieeexplore.ieee.org/document/10460524/ |
| work_keys_str_mv | AT karimasattar roadcrashinjuryseveritypredictionusingagraphneuralnetworkframework AT iskandarishak roadcrashinjuryseveritypredictionusingagraphneuralnetworkframework AT lillysurianiaffendey roadcrashinjuryseveritypredictionusingagraphneuralnetworkframework AT sitinurulainbintimohdrum roadcrashinjuryseveritypredictionusingagraphneuralnetworkframework |