Ambrose Mohandoss, Melvin Akash (2023) Implementing Machine Learning Models for Predicting Road Accident Severity in Northern Ireland. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (2MB) | Preview |
Abstract
This research project addresses the challenge of enhancing road safety in Northern Ireland by employing advanced machine learning and deep learning techniques to predict road accident severity. By evaluating various algorithms, the project focuses on hyperparameter tuning, a critical step in optimizing the performance of machine learning models. The tuning process is driven by the goal of achieving the highest possible accuracy in predicting accident severity, which is crucial for developing effective model building. The research successfully identifies Random Forest and K-Nearest Neighbors (KNN) as the most effective models, with remarkably high accuracy rates of 98.49% and 99.01% respectively. However, it also uncovers limitations within the artificial Neural Network (ANN) model, indicating a potential area for further refinement. The findings of this study provide valuable insights that can guide policy-making and the design of targeted road safety interventions, potentially saving lives and reducing the frequency and severity of road accidents.
Item Type: | Thesis (Masters) |
---|---|
Supervisors: | Name Email Sahni, Anu UNSPECIFIED |
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: | Ciara O'Brien |
Date Deposited: | 06 May 2025 18:07 |
Last Modified: | 06 May 2025 18:07 |
URI: | https://norma.ncirl.ie/id/eprint/7491 |
Actions (login required)
![]() |
View Item |