Towards an Accident Severity Prediction System with Logistic Regression

Sie sind hier: Homepage > Suche

Towards an Accident Severity Prediction System with Logistic Regression

Mensouri, Houssam / Azmani, Abdellah / Azmani, Monir

The main negative impacts of road accidents are injuries and deaths. According to statistics from the World Health Organization (WHO), traffic accidents resulted in the deaths of about 1.35 million people in 2016 worldwide. Collisions result in property damage as well as the death of road users. Black spots are the locations on a road where accidents are common or the places where collisions result in fatal injuries. Therefore, predicting the accident severity is considered one of the most effective approaches to avoid any damage.

This article focuses on predicting the severity of traffic accidents. The model was built on a database that contains more than 4 million total accidents that occurred from February 2016 to December 2020 nationwide, covering 49 states in the United States. Traffic accident data was used to build a classifier based on an artificial intelligence algorithm (Logistic Regression). Our model is then evaluated using the different evaluation criteria (confusion matrix, accuracy, precision, specificity, F1-score). The experimental results revealed that the logistic regression model classifier can predict the severity of accidents with an accuracy equal to 88%.

Zugriff

Zugriff prüfen

Verfügbarkeit in meiner Bibliothek prüfen

Bestellung bei Subito €

Exportieren, teilen und zitieren

Dokumentinformationen

Titel :

Towards an Accident Severity Prediction System with Logistic Regression

Weitere Titelangaben:

Lect. Notes in Networks, Syst.

Beteiligte:

Kacprzyk, Janusz (Herausgeber:in) / Ezziyyani, Mostafa (Herausgeber:in) / Balas, Valentina Emilia (Herausgeber:in) / Mensouri, Houssam (Autor:in) / Azmani, Abdellah (Autor:in) / Azmani, Monir (Autor:in)

Kongress:

International Conference on Advanced Intelligent Systems for Sustainable Development ; 2022 ; Rabat, Morocco May 22, 2022 - May 27, 2022

Erschienen in:

International Conference on Advanced Intelligent Systems for Sustainable Development ; Kapitel : 34 ; 396-410

Lecture Notes in Networks and Systems ; 637

Erscheinungsdatum :

2023-06-10

Format / Umfang :

15 pages

ISBN :

978-3-031-26384-2 , 978-3-031-26383-5

ISSN :