Traffic Violation Arrests Using Machine Learning Approaches

Shephali, Akanksha (2024) Traffic Violation Arrests Using Machine Learning Approaches. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (889kB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Offenses related to the roads are deemed a huge problem that affects public safety as well as the systems of using the roads. Traffic violation arrest prediction can help improve the implementation of the law due to proper planning and utilization of available resources. Conventional solutions to this problem involve either using people to go through the data sets and trying to identify the relevant patterns manually or designing a set of rules that can do the same thing but is not scalable. In this paper, different ML techniques will be discussed and examined such as Random Forest, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression, and Linear Regression. Deep Neural Decision Forest, Multilayer perceptron, Recurrent neural network (RNN), LSTM, and Gated Recurrent Unit (GRU) are used as the deep learning approaches. Thus, in the data preprocessing, feature scaling was applied, and categorical features were encoded by using one hot encoding. Confusion matrix, accuracy, precision, recall, F1 score, and AUC ROC charts have been adopted with variance. Among the above-said ML models, Random Forest outperformed all the other models with an accuracy of 88% and the highest Recall and F1 Score of 80.21% and 34.15% respectively. However, GRU performed best among the deep learning models with an accuracy of 96.14% and Recall and F1 Score being 34.04% and 40.44%. The results presented here demonstrate that both types of models can help predict arrests for traffic violations which could aid police work by predicting when an arrest is likely to occur though it’s important to balance accuracy with transparency.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Rustam, Furqan 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:	Ciara O'Brien
Date Deposited:	05 Sep 2025 10:24
Last Modified:	05 Sep 2025 10:24
URI:	https://norma.ncirl.ie/id/eprint/8811

Actions (login required)

View Item