Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network
An inherent feature of road accident databases is the imbalance between the number of observations associated with accidents with fatal and non-fatal victims of injuries concerning to accidents without victims. This particularity led to the adoption of corresponding balancing techniques, which can r...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Associação Nacional de Pesquisa e Ensino em Transportes (ANPET)
2020-12-01
|
Series: | Transportes |
Subjects: | |
Online Access: | https://www.revistatransportes.org.br/anpet/article/view/2271 |
id |
doaj-4baa3a329d664a0fa660b0e1a4ee1887 |
---|---|
record_format |
Article |
spelling |
doaj-4baa3a329d664a0fa660b0e1a4ee18872021-05-13T20:06:37ZengAssociação Nacional de Pesquisa e Ensino em Transportes (ANPET)Transportes2237-13462020-12-0128510.14295/transportes.v28i5.2271Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural NetworkMaria Lígia Chuerubim0Leonardo N. Ferreira1Alan D.B. Valejo2Bárbara Stolte Bezerra3Giuliano Sant'Anna Marotta4Irineu da Silva5Faculty of Civil Engineering, Federal University of Uberlândia, Brazil.Associate Laboratory of Computation and Applied Mathematics National Institute of Space Research (INPE), BrazilInstitute of Mathematical and Computer Sciences, School of Engineering of São Carlos, University of São Paulo, Brazil.Faculty of Civil Engineering, UNESP Sao Paulo State University, Brazil.Faculty of Civil Engineering, Federal University of Uberlândia, Brazil.Department of Transport Engineering, School of Engineering of São Carlos, University of São Paulo, Brazil.An inherent feature of road accident databases is the imbalance between the number of observations associated with accidents with fatal and non-fatal victims of injuries concerning to accidents without victims. This particularity led to the adoption of corresponding balancing techniques, which can resample classes and attributes. Therefore, it ensures that there is no over-adjustment of the data in classification problems. This study investigates the influence of different balancing methods such as undersampling, oversampling and SMOTE on the classification process of road accident severity adopting an Artificial Neural Network approach. The results obtained indicate that all methods used were able to effectively adjust the balance between the minority and majority classes. Balancing leads to a better performance of the classifier, shown by the efficient adjustment of the data to the model, as the gain in the quality and accuracy of the classification process, especially, considering sampling techniques such as SMOTE. https://www.revistatransportes.org.br/anpet/article/view/2271Imbalanced data. Accident severity. Classification and Artificial Neural Networks. |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Maria Lígia Chuerubim Leonardo N. Ferreira Alan D.B. Valejo Bárbara Stolte Bezerra Giuliano Sant'Anna Marotta Irineu da Silva |
spellingShingle |
Maria Lígia Chuerubim Leonardo N. Ferreira Alan D.B. Valejo Bárbara Stolte Bezerra Giuliano Sant'Anna Marotta Irineu da Silva Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network Transportes Imbalanced data. Accident severity. Classification and Artificial Neural Networks. |
author_facet |
Maria Lígia Chuerubim Leonardo N. Ferreira Alan D.B. Valejo Bárbara Stolte Bezerra Giuliano Sant'Anna Marotta Irineu da Silva |
author_sort |
Maria Lígia Chuerubim |
title |
Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network |
title_short |
Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network |
title_full |
Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network |
title_fullStr |
Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network |
title_full_unstemmed |
Evaluation of database balancing techniques for road accident severity classification employing Artificial Neural Network |
title_sort |
evaluation of database balancing techniques for road accident severity classification employing artificial neural network |
publisher |
Associação Nacional de Pesquisa e Ensino em Transportes (ANPET) |
series |
Transportes |
issn |
2237-1346 |
publishDate |
2020-12-01 |
description |
An inherent feature of road accident databases is the imbalance between the number of observations associated with accidents with fatal and non-fatal victims of injuries concerning to accidents without victims. This particularity led to the adoption of corresponding balancing techniques, which can resample classes and attributes. Therefore, it ensures that there is no over-adjustment of the data in classification problems. This study investigates the influence of different balancing methods such as undersampling, oversampling and SMOTE on the classification process of road accident severity adopting an Artificial Neural Network approach. The results obtained indicate that all methods used were able to effectively adjust the balance between the minority and majority classes. Balancing leads to a better performance of the classifier, shown by the efficient adjustment of the data to the model, as the gain in the quality and accuracy of the classification process, especially, considering sampling techniques such as SMOTE.
|
topic |
Imbalanced data. Accident severity. Classification and Artificial Neural Networks. |
url |
https://www.revistatransportes.org.br/anpet/article/view/2271 |
work_keys_str_mv |
AT marialigiachuerubim evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork AT leonardonferreira evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork AT alandbvalejo evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork AT barbarastoltebezerra evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork AT giulianosantannamarotta evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork AT irineudasilva evaluationofdatabasebalancingtechniquesforroadaccidentseverityclassificationemployingartificialneuralnetwork |
_version_ |
1721441860547772416 |