No abstract

Crime rate tends to be on the increase across the globe, and crime data analysis becomes imperative to aid predictive policing in tackling incidence of crime. In this paper data mining approach was applied to violent crime dataset for predicting next occurrence of violent crime. Previous researchers have used different supervised learning algorithms for crime prediction with accuracy results left to be improved upon. Consequently, this study particularly apply decision tree C5.0 algorithm on violent crime dataset in order to determine the probability of next occurrence of violent crime in Lagos metropolis. The data used was derived from Nigerian Police statistic department Obalende Lagos, pre-processed and applied on decision tree model built. The model was evaluated using the six violent crime types (murder, arm robbery, kidnapping, rape, non-negligent assault and man slaughter) dataset. The results obtained were evaluated using confusion matric and found to return an accuracy of 76.4% (percent). Based on this result, the model could be used by the Police authority to strategize and plan towards mitigating crime rate in the country.

No abstract

No abstract