The Role of Artificial Intelligence in Predicting and Preventing Crime in Nigeria

Abstract

Artificial intelligence (AI) presents transformative potential for crime prediction and prevention in Nigeria, a country characterized by multidimensional insecurity, institutional capacity deficits, and a rapidly expanding digital infrastructure. This study evaluated the performance of six machine learning algorithms—Random Forest, Support Vector Machine, Long Short-Term Memory Neural Networks, Logistic Regression, XGBoost Gradient Boosting, and Naïve Bayes—applied to crime prediction tasks using Nigerian crime datasets from 2016 to 2022. Pilot deployments of AI-assisted predictive policing in Lagos, Kano, the FCT, Rivers, and Ogun States were assessed using pre-post quasi-experimental design. The LSTM Neural Network recorded the highest prediction accuracy (91.2%) and AUC-ROC score (0.954), while the FCT Abuja deployment achieved the largest crime reduction (25.8% over 24 months). A stakeholder survey (n = 214) identified inadequate funding (91.4%), infrastructure limitations (88.1%), and poor data quality (84.2%) as the primary barriers to AI adoption in Nigerian law enforcement. The study concludes that AI-assisted predictive policing holds significant empirical promise for Nigeria but requires an enabling ecosystem of data governance, institutional capacity, ethical oversight, and sustained financing. A phased National AI Crime Prevention Strategy is proposed.