A STUDY ON SOFTWARE RELIABILITY EVALUATION BASED ON SOFTWARE METRICS DATA USING RBF NETWORKS

This paper provides a new approach to estimate the software reliability with the data sets obtained from the upper process of software development. We first explain the concept of RBF (Radial Basis Function) network and show the structure of our model. By using the actual data sets, we try to estimate the number of inherent software faults. We also compare the proposed model and the neural network model which were constructed in the previous study in terms of the goodness-of-fit to the data sets. Our model can provide the better estimation results of the number of inherent software faults in the early stage of software development