NORMA eResearch @NCI Library

A Machine Learning Based Comparative Analysis of Accident Severity Prediction Mechanism in USA

Mohire, Soham Padmakar (2022) A Machine Learning Based Comparative Analysis of Accident Severity Prediction Mechanism in USA. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (3MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


Traffic accidents that occur worldwide are a concerning issue as it results in major deaths and injuries. This burden of casualties tends to be higher in developing countries. Hence, a model to predict the occurrence of accidents is a significant challenge. However, one of the substantial ways to predict the severity of such accidents is through the implementation of machine learning algorithms. Therefore, the primary aim of the proposed thesis is to automate the process of accident detection by analysing the levels of severity and filtering a set of impactful factors that might result into a road accident. For this purpose, a dataset from Kaggle repository is obtained that contains a countrywide car accident data of USA from Dec 2016 to Dec 2021. Theoretical concepts of SMOTE is implemented to balance the dataset and thereby handle data imbalance. Later, the dataset is used to develop a framework based on four machine learning algorithms and one stacking algorithm. Finally, an analysis is done based on factors such as weather conditions and different severity levels that might lead to the occurrence of road accidents. The experimental analysis so conducted in the study indicates that the random forest model has performed better in comparison to all the implemented models by generating an accuracy of 74 percent.

Item Type: Thesis (Masters)
Milosavljevic, Vladimir
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > TL Motor vehicles. Aeronautics. Astronautics
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 23 May 2023 15:00
Last Modified: 23 May 2023 15:04

Actions (login required)

View Item View Item