Severity prediction of traffic accidents with recurrent neural networks
In this paper, a deep learning model using a Recurrent Neural Network (RNN) was developed and employed to predict the injury severity of traffic accidents based on 1130 accident records that have occurred on the North-South Expressway (NSE), Malaysia over a six-year period from 2009 to 2015. Compare...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI
2017
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/63203/1/Severity%20Prediction%20of%20Traffic%20Accidents%20with%20Recurrent%20Neural%20Networks.pdf |
| _version_ | 1796977704912814080 |
|---|---|
| author | Sameen, Maher Ibrahim Pradhan, Biswajeet |
| author_facet | Sameen, Maher Ibrahim Pradhan, Biswajeet |
| author_sort | Sameen, Maher Ibrahim |
| collection | UPM |
| description | In this paper, a deep learning model using a Recurrent Neural Network (RNN) was developed and employed to predict the injury severity of traffic accidents based on 1130 accident records that have occurred on the North-South Expressway (NSE), Malaysia over a six-year period from 2009 to 2015. Compared to traditional Neural Networks (NNs), the RNN method is more effective for sequential data, and is expected to capture temporal correlations among the traffic accident records. Several network architectures and configurations were tested through a systematic grid search to determine an optimal network for predicting the injury severity of traffic accidents. The selected network architecture comprised of a Long-Short Term Memory (LSTM) layer, two fully-connected (dense) layers and a Softmax layer. Next, to avoid over-fitting, the dropout technique with a probability of 0.3 was applied. Further, the network was trained with a Stochastic Gradient Descent (SGD) algorithm (learning rate = 0.01) in the Tensorflow framework. A sensitivity analysis of the RNN model was further conducted to determine these factors’ impact on injury severity outcomes. Also, the proposed RNN model was compared with Multilayer Perceptron (MLP) and Bayesian Logistic Regression (BLR) models to understand its advantages and limitations. The results of the comparative analyses showed that the RNN model outperformed the MLP and BLR models. The validation accuracy of the RNN model was 71.77%, whereas the MLP and BLR models achieved 65.48% and 58.30% respectively. The findings of this study indicate that the RNN model, in deep learning frameworks, can be a promising tool for predicting the injury severity of traffic accidents. |
| first_indexed | 2024-03-06T09:43:58Z |
| format | Article |
| id | upm.eprints-63203 |
| institution | Universiti Putra Malaysia |
| language | English |
| last_indexed | 2024-03-06T09:43:58Z |
| publishDate | 2017 |
| publisher | MDPI |
| record_format | dspace |
| spelling | upm.eprints-632032018-08-20T06:29:50Z http://psasir.upm.edu.my/id/eprint/63203/ Severity prediction of traffic accidents with recurrent neural networks Sameen, Maher Ibrahim Pradhan, Biswajeet In this paper, a deep learning model using a Recurrent Neural Network (RNN) was developed and employed to predict the injury severity of traffic accidents based on 1130 accident records that have occurred on the North-South Expressway (NSE), Malaysia over a six-year period from 2009 to 2015. Compared to traditional Neural Networks (NNs), the RNN method is more effective for sequential data, and is expected to capture temporal correlations among the traffic accident records. Several network architectures and configurations were tested through a systematic grid search to determine an optimal network for predicting the injury severity of traffic accidents. The selected network architecture comprised of a Long-Short Term Memory (LSTM) layer, two fully-connected (dense) layers and a Softmax layer. Next, to avoid over-fitting, the dropout technique with a probability of 0.3 was applied. Further, the network was trained with a Stochastic Gradient Descent (SGD) algorithm (learning rate = 0.01) in the Tensorflow framework. A sensitivity analysis of the RNN model was further conducted to determine these factors’ impact on injury severity outcomes. Also, the proposed RNN model was compared with Multilayer Perceptron (MLP) and Bayesian Logistic Regression (BLR) models to understand its advantages and limitations. The results of the comparative analyses showed that the RNN model outperformed the MLP and BLR models. The validation accuracy of the RNN model was 71.77%, whereas the MLP and BLR models achieved 65.48% and 58.30% respectively. The findings of this study indicate that the RNN model, in deep learning frameworks, can be a promising tool for predicting the injury severity of traffic accidents. MDPI 2017 Article PeerReviewed text en http://psasir.upm.edu.my/id/eprint/63203/1/Severity%20Prediction%20of%20Traffic%20Accidents%20with%20Recurrent%20Neural%20Networks.pdf Sameen, Maher Ibrahim and Pradhan, Biswajeet (2017) Severity prediction of traffic accidents with recurrent neural networks. Applied Sciences, 7 (6). pp. 1-17. ISSN 2076-3417 10.3390/app7060476 |
| spellingShingle | Sameen, Maher Ibrahim Pradhan, Biswajeet Severity prediction of traffic accidents with recurrent neural networks |
| title | Severity prediction of traffic accidents with recurrent neural networks |
| title_full | Severity prediction of traffic accidents with recurrent neural networks |
| title_fullStr | Severity prediction of traffic accidents with recurrent neural networks |
| title_full_unstemmed | Severity prediction of traffic accidents with recurrent neural networks |
| title_short | Severity prediction of traffic accidents with recurrent neural networks |
| title_sort | severity prediction of traffic accidents with recurrent neural networks |
| url | http://psasir.upm.edu.my/id/eprint/63203/1/Severity%20Prediction%20of%20Traffic%20Accidents%20with%20Recurrent%20Neural%20Networks.pdf |
| work_keys_str_mv | AT sameenmaheribrahim severitypredictionoftrafficaccidentswithrecurrentneuralnetworks AT pradhanbiswajeet severitypredictionoftrafficaccidentswithrecurrentneuralnetworks |