Software Reliability Model Estimation for an Indeterministic Crime Cluster through Reinforcement Learning

Abstract

The software reliability model estimates the probability of data failure in a specific environment, significantly impacting reliability and trustworthiness. The paper study focuses on cluster crime data, i.e., indeterministic in Neutrosophic Logic, using a software reliability model. The study utilizes reinforcement learning, Neutrosophic logic, and non-homogeneous Poisson process crime data to estimate indeterministic cluster data in crime. The "Non-homogeneous Poisson Process with Neutrosophic Logic" technique performs well in evaluating and deterring crime based on crime data analysis. The crime cluster involving offenders correctly classified as failure to accomplish does better than uncertain cluster reliability estimation with least squares and logistic regression analysis. The method enables crime prediction and prevention by using concave growth models to create an uncertain crime cluster, penalizing the correct person.