The important characteristics of a software quality assurance system is software reliability. It is the probability of error-free software process in a given environment for a given interval. A huge number of research projects have been attempted to increase the software’s reliability. Software modelling, software measurement, and software improvement are a three-step process for enhancing software reliability. Decision Trees (DT), Support Vector Machines (SVM), Artificial Neural Networks (ANN), and other Machine Learning (ML) techniques for forecasting software reliability are reviewed in this study. Software Reliability is a measure for the success whether the software is functioning as per expectations in a given time (interval or point) in the environment that is prevailing. The purpose of ML methods to forecast software stability has yielded careful and impressive findings. To identify the value of every approach in assessing the competence of software dependability prediction models, a comparative study is also undertaken. In this review paper, several methods of Software Reliability and outcomes using ML are explained and reviewed.