Estimate reliability of component-based software sys-tem using modified neuro fuzzy model

Abstract

There are many algorithms and techniques for estimating the reliability of Component Based Software Systems (CBSSs). Accurate estimation depends on two factors: component reliability and glue code reliability. Still much more research is expected to estimate reliability in a better way. A number of soft computing approaches for estimating CBSS reliability has been proposed. These techniques learnt from the past and capture existing patterns in data. In this paper, we proposed new model for estimating CBSS reliability known as Modified Neuro Fuzzy Inference System (MNFIS). This model is based on four factors Reusability, Operational, Component dependency, Fault Density. We analyze the proposed model for diffent data sets and also compare its performance with that of plain Fuzzy Inference System. Our experimental results show that, the proposed model gives better reliability as compare to FIS.

Software Reliability Assessment Using Hybrid Neuro-Fuzzy Model

“…In this section, identification of reliability-relevant metrics was made. As failure data is not available at the early phase to predict reliability, identified metrics named Complexity, Changeability, and Portability are used to develop a model for the early detection of software reliability [25]. The linguistic values from all these three identified metrics related to reliability were considered input to the proposed model to predict the reliability of the software in the early stage of software development.…”