Prediction of Crime Rate in Diverse Environs Using Hybrid Classifier

Abstract

Crime is the fear and terror among the populace worldwide. Crime is an inherent component of the hazards we encounter daily. In recent times, the mass media has extensively covered numerous criminal incidents, including theft, rape and sexual offenses, robbery, murder, and kidnappings. Various works have been produced to understand the factors that lead to an individual committing a criminal act, the potential dangers involved, and strategies to prevent it. The crime computation technique aims to forecast crime rates, enabling police officers to avoid such incidents effectively. Based on this, a novel prediction approach utilizing a hybrid classifier is suggested. An evaluation of the proposed method was conducted using several criteria. The performance of this recently constructed hybrid prediction model is evaluated by comparing it with established models such as Genetic Algorithm, Particle Swarm Optimization, and Firefly Algorithm. Various performance measures, including error rate, sensitivity, specificity, precision, and execution time, are used for the comparison. Based on the results, this hybrid model is the most optimal crime prediction model compared to the other current models. The suggested approach is executed using the JAVA programming language.