According to the Office for National Statistics (2023), in the year ending September 2022, adults aged 16 years and over experienced 9.1 million offences. These high crime rates can be significant and far-reaching, affecting individuals and society in the short and long term. Machine learning techniques can be implemented to predict crimes that may occur in the
future based on information such as the area, date and time. However, it is not clear which approach is best. Therefore, this project aims to evaluate different machine learning approaches to find the best way to predict crime accurately. Four machine learning classifiers will be compared: Naive Bayes, Decision Tree, Naive Bayes and K-Nearest Neighbours. The
use of K-Fold Cross Validation and Hyperparameter Tuning will also be measured based on a variety of performance metrics. The findings show that using Hyperparameter Tuning is necessary to yield better results, and the Random Forest classifier with Hyperparameter Tuning yields the best results.
PLEASE NOTE: You must be a member of the University of Lincoln to be able to view this dissertation. Please log in here.