1
|
Li Y, Yang Z, Xing L, Yuan C, Liu F, Wu D, Yang H. Crash injury severity prediction considering data imbalance: A Wasserstein generative adversarial network with gradient penalty approach. ACCIDENT; ANALYSIS AND PREVENTION 2023; 192:107271. [PMID: 37659275 DOI: 10.1016/j.aap.2023.107271] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/28/2022] [Revised: 07/29/2023] [Accepted: 08/24/2023] [Indexed: 09/04/2023]
Abstract
For each road crash event, it is necessary to predict its injury severity. However, predicting crash injury severity with the imbalanced data frequently results in ineffective classifier. Due to the rarity of severe injuries in road traffic crashes, the crash data is extremely imbalanced among injury severity classes, making it challenging to the training of prediction models. To achieve interclass balance, it is possible to generate certain minority class samples using data augmentation techniques. Aiming to address the imbalance issue of crash injury severity data, this study applies a novel deep learning method, the Wasserstein generative adversarial network with gradient penalty (WGAN-GP), to investigate a massive amount of crash data, which can generate synthetic injury severity data linked to traffic crashes to rebalance the dataset. To evaluate the effectiveness of the WGAN-GP model, we systematically compare performances of various commonly-used sampling techniques (random under-sampling, random over-sampling, synthetic minority over-sampling technique and adaptive synthetic sampling) with respect to dataset balance and crash injury severity prediction. After rebalancing the dataset, this study categorizes the crash injury severity using logistic regression, multilayer perceptron, random forest, AdaBoost and XGBoost. The AUC, specificity and sensitivity are employed as evaluation indicators to compare the prediction performances. Results demonstrate that sampling techniques can considerably improve the prediction performance of minority classes in an imbalanced dataset, and the combination of XGBoost and WGAN-GP performs best with an AUC of 0.794 and a sensitivity of 0.698. Finally, the interpretability of the model is improved by the explainable machine learning technique SHAP (SHapley Additive exPlanation), allowing for a deeper understanding of the effects of each variable on crash injury severity. Findings of this study shed light on the prediction of crash injury severity with data imbalance using data-driven approaches.
Collapse
Affiliation(s)
- Ye Li
- School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China; Hunan Key Laboratory of Smart Roadway and Cooperative Vehicle-Infrastructure Systems, Changsha University of Science & Technology, Changsha, 410114 Hunan, China.
| | - Zhanhao Yang
- School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China.
| | - Lu Xing
- School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha, Hunan 410114, China.
| | - Chen Yuan
- School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China; Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong, China.
| | - Fei Liu
- School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China.
| | - Dan Wu
- School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan 410075, China.
| | - Haifei Yang
- School of Civil and Transportation Engineering, Hohai University, Nanjing, Jiangsu 210098, China.
| |
Collapse
|
2
|
Analysis of Crash Severity of Texas Two Lane Rural Roads Using Solar Altitude Angle Based Lighting Condition. SUSTAINABILITY 2022. [DOI: 10.3390/su14031692] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/10/2022]
Abstract
Many studies have examined the impact of factors affecting accident severity in rural areas; however, little attention has been paid to different lighting conditions (LCs), and less to the detailed categories and precise determining of twilight. In this paper, solar altitude angle (SAA), as a basis for differentiating and categorizing LCs, is proposed to investigate explanatory variables in much greater detail. For each LC, namely, dark, twilight, dark lit (dark with street lights) and daylight, separate random parameter models are developed to investigate the impacts of some factors on crash injury severity data of 2017 and 2018 in two lane rural roads of Texas. The model estimation results indicated that different LCs have various contributing factors, indeed, to each injury severity, further stressing the significance of investigating crashes based on SAA. The key differences include crash location, marked lane, grade direction, no passing zone, shoulder width, weekday and collision type. The important findings were that developing artificial lighting at intersections and LED raised pavement markers on two lane rural roads could lead to enhanced road safety under dark LCs. Furthermore, increasing shoulder width in straight segments of two lane rural roads is important for decreasing severe injury in daylight conditions.
Collapse
|
3
|
Islam Z, Abdel-Aty M, Cai Q, Yuan J. Crash data augmentation using variational autoencoder. ACCIDENT; ANALYSIS AND PREVENTION 2021; 151:105950. [PMID: 33370603 DOI: 10.1016/j.aap.2020.105950] [Citation(s) in RCA: 25] [Impact Index Per Article: 8.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/28/2020] [Revised: 12/06/2020] [Accepted: 12/10/2020] [Indexed: 06/12/2023]
Abstract
In this paper, we present a data augmentation technique to reproduce crash data. The dataset comprising crash and non-crash events are extremely imbalanced. For instance, the dataset used in this paper consists of only 625 crash events for over 6.5 million non-crash events. Thus, learning algorithms tend to perform poorly on these datasets. We have used variational autoencoder to encode all the events into a latent space. After training, the model could successfully separate crash and non-crash events. To generate data, we sampled from the latent space containing crash data. The generated data was compared with the real data from different statistical aspects. t-Test, Levene-test and Kolmogrove Smirnov test showed that the generated data was statistically similar to the real data. It was also compared to some of the minority oversampling techniques like SMOTE and ADASYN as well as the GAN framework for generating data. Crash prediction models based on Logistic Regression (LR), Support Vector Machine (SVM) and Artificial Neural Network (ANN) were used to compare the generated data from the different oversampling techniques. Overall, variational autoencoder (VAE) showed excellent results compared to the other data augmentation methods. Specificity is improved by 8% and 4% for VAE-LR and VAE-SVM respectively when compared to SMOTE while the sensitivity is improved by 6% and 5% when compared to ADASYN. Moreover, VAE generated data also helps to overcome the overfitting problem in SMOTE and ADASYN since there is flexibility in choosing the decision boundary.
Collapse
Affiliation(s)
- Zubayer Islam
- Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL, 32816, USA.
| | - Mohamed Abdel-Aty
- Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL, 32816, USA.
| | - Qing Cai
- Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL, 32816, USA.
| | - Jinghui Yuan
- Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL, 32816, USA.
| |
Collapse
|
4
|
Influential Factors on Injury Severity for Drivers of Light Trucks and Vans with Machine Learning Methods. SUSTAINABILITY 2020. [DOI: 10.3390/su12041324] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
The study of road accidents and the adoption of measures to reduce them is one of the most important targets of the Sustainable Development Goals for 2030. To further progress in the improvement of road safety, it is necessary to focus studies on specific groups, such as light trucks and vans. Since 2013 in Spain, there has been an upturn in accidents in these two categories of vehicles and a renewed interest to deepen our understanding of the causes that encourage this behavior. This paper focuses on using machine learning methods to explain driver-injury severity in run-off-roadway and rollover types of accidents. A Random Forest (RF)-classification tree (CART) approach is used to select the relevant categorical variables (driver, vehicle, infrastructure, and environmental factors) to obtain models that classify, explain, and predict the severity of such accidents with good accuracy. A support vector machine and binomial logit models were applied in order to contrast the variable importance ranking and the performance analysis, and the results are convergent with the RF+CART approach (more than 70% accuracy). The resulting models highlight the importance of using safety belts, as well as psychophysical conditions (alcohol, drugs, or sleep deprivation) and injury localization for the two accident types.
Collapse
|