Modified Artificial Neural Networks in Stochastic Differential Equations to Estimate the Occurrence Rate of a New Process for Software Reliability Growth Modelling

This paper aims to utilize modified artificial neural networks under stochastic differential equations to improve the estimation of occurrence rates on models towards the growth of software reliability through such models. There has never been a time when reliable assessment methods have been more important as software systems grow ever more sophisticated. This study covers the risks arising from software bugs, security weaknesses, and the need to meet legal requirements. The authors use a new method that combines artificial intelligence techniques with classical reliability models to achieve their goal of a more efficient computational tool for predicting software reliability. It focuses on the assessment of the differentiation of the Goel Okumoto model for the reliability functions and then simulates the event rate to obtain further applications. The implications of this research are vast because it provides a sound method through which time to failure in crucial systems essential in electricity generation, such as reliability engineering, finance, and telecommunication, is forecasted. In the end, this research benefits the existing conversations on enhancing software quality and reliability as organizations strive to provide organizations with robust and secure software solutions to handle today’s advanced technologies.