ACCIDENT SEVERITY PREDICTION FOR ROAD TRANSPORTATION SYSTEM USING MACHINE LEARNING
Fuente:
Zenodo
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Recurso digital |
| Pubblicazione: |
Zenodo
2026
|
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866901230249312256 |
|---|---|
| author | UMEJURU DANIEL IGBUDU KINGSLEY EZEBUNWO |
| author_facet | UMEJURU DANIEL IGBUDU KINGSLEY EZEBUNWO |
| contents | <p><em><span>Road accidents constitute some of the most horrifying experience one can have and can sometimes leave an indelible mark on victims for the rest of their life. The constant occurrence of accidents involving cars and collisions is an immediate consequence of road conditions, related environmental factors, and the carelessness of drivers involved.<span> </span>Professionals and agencies can utilize accident severity prediction models to obtain insight into the factors influencing road traffic incidents, and this can help predict the degree of severity of traffic accidents. Machine learning algorithms can be used to determine accident trends and predict cases of fatalities, major injuries, or minor injuries. The aim is to provide a method for predicting the degree of severity caused by a traffic accident using a ML algorithm. We created a prediction model using support vector machine (SVM). Road traffic accident (RTA) datasets obtained from the Kaggle website served as the data in the experiment. The artificial neural network was designed with the decay concept in mind, which penalizes the model to acquire knowledge from larger weights and forces it to learn from smaller weights, leading to a simpler architecture. The fine-tuned hyper-parameter variables enhanced the classification results and assisted in minimizing the model's complexity with the aid of the penalty term. The SVM yielded a recommendable result of at 89.0% efficiency which shows that the model performed excellently as expected</span></em></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18270574 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | ACCIDENT SEVERITY PREDICTION FOR ROAD TRANSPORTATION SYSTEM USING MACHINE LEARNING UMEJURU DANIEL IGBUDU KINGSLEY EZEBUNWO <p><em><span>Road accidents constitute some of the most horrifying experience one can have and can sometimes leave an indelible mark on victims for the rest of their life. The constant occurrence of accidents involving cars and collisions is an immediate consequence of road conditions, related environmental factors, and the carelessness of drivers involved.<span> </span>Professionals and agencies can utilize accident severity prediction models to obtain insight into the factors influencing road traffic incidents, and this can help predict the degree of severity of traffic accidents. Machine learning algorithms can be used to determine accident trends and predict cases of fatalities, major injuries, or minor injuries. The aim is to provide a method for predicting the degree of severity caused by a traffic accident using a ML algorithm. We created a prediction model using support vector machine (SVM). Road traffic accident (RTA) datasets obtained from the Kaggle website served as the data in the experiment. The artificial neural network was designed with the decay concept in mind, which penalizes the model to acquire knowledge from larger weights and forces it to learn from smaller weights, leading to a simpler architecture. The fine-tuned hyper-parameter variables enhanced the classification results and assisted in minimizing the model's complexity with the aid of the penalty term. The SVM yielded a recommendable result of at 89.0% efficiency which shows that the model performed excellently as expected</span></em></p> |
| title | ACCIDENT SEVERITY PREDICTION FOR ROAD TRANSPORTATION SYSTEM USING MACHINE LEARNING |
| url | https://doi.org/10.5281/zenodo.18270574 |