A Survey on Analysis and Prediction of Crime Hotspots using Stacked Generalization Approach

Ensemble learning is a collaborative decision-making process that aggregates the predictions of learnt classifiers to generate new examples. This paper suggests a practical, real-world approach called the assemble-stacking based crime prediction technique (SBCPM), which is based on algorithms for making accurate forecasts of crime. In comparison to past investigations, the proposed method showed a better predictive potential and yielded classification findings that were more accurate when applied to the testing data. The recommended method suggests that the ensemble model’s prediction accuracy is higher than that of the individual classifiers, which is helpful for making prospective predictions about crimes.