Sexual Crime Prediction in an African Context

Abstract

The growth of sexual crime in Africa and the incapacity to control it have had a major physical and psychological impact on victims. Crime in general can reduce foreign direct investment in a country. This study was driven by the need to reduce sexual crime across the country. Data mining techniques were applied to a sexual crime dataset extracted from the South African crime statistics database on Kaggle to visualise sexual crime trends and create a model that predicts the occurrence of sexual crimes, thereby helping government and law enforcement agencies gain insights into the most common sexual crime hotspots across all the nine provinces of South Africa. This model may help law enforcement combat sexual crimes faster. We identified data analytics methods for sexual crime prediction and chose the best one. Because of a linear relationship between the dependent variable (sexual crime) and the independent variables (population and density), linear regression and decision tree classifier algorithms were used to predict main causes of sexual crime in South Africa. Accuracy, precision, recall, and F1 scores were used to test the decision tree algorithm's performance. Linear regression was measured using the R-squared score, which achieved 91% accuracy, indicating how well the model will predict sexual crime occurrences.