Comparative Study of Machine Learning and Deep Learning Approaches for Predicting Irish Road Accidents

Rajan, Karthika (2023) Comparative Study of Machine Learning and Deep Learning Approaches for Predicting Irish Road Accidents. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (1MB) | Preview

Abstract

In past few years, the crashes on roads in Ireland have become an increasing concern. According to Road Safety Authority of Ireland (RSA), number of crashes in Ireland has increased by 13 percentage, demonstrating the need for further measures for lowering the crash rate. The purpose of this research is to perform a comparison between several machine learning and deep learning algorithms on accident dataset of Ireland in order to determine the ideal technique for accident forecasting. This project also intends to determine if hyperparameter optimisation can improve the performance of the ML techniques. The evaluated machine learning algorithms are Random Forest, Decision Tree, XGBoost, Ridge Regression and KNN, whereas the deep learning models are Long Short-Term Memory (LSTM) and Feedforward Neural Network (FNN). Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and R-squared score are used to measure the effectiveness of the models. The results indicate that the hyperparameter optimisation enhanced the performance of the machine learning models significantly. After tuning their hyperparameters, XGBoost outperformed other models followed by Random Forest and Ridge Regression. XGBoost has got an R-squared score of 0.96. KNN algorithm underperformed when compared to other models. For deep learning models, FNN model performed better than LSTM algorithm. This study offers important insights into the application of machine learning and deep learning models for prediction of traffic accidents in Ireland. The results can definitely help the policymakers and the road safety authorities to allocate resources more efficiently. By this it is possible to reduce the accidents and can enhance road safety in the nation.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Syed, Muslim Jameel UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TE Highway engineering. Roads and pavements 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:	28 Dec 2024 15:00
Last Modified:	28 Dec 2024 15:00
URI:	https://norma.ncirl.ie/id/eprint/7252

Actions (login required)

View Item