Data-driven software reliability evaluation under incomplete knowledge on fault count distribution

In this article, we consider data-driven approaches for software reliability evaluation without specifying the fault count distribution, where the underlying stochastic process to describe software fault-counts in the system testing is given by a non-homogeneous Poisson process. A comprehensive non-parametric method based on the kernel estimation is provided with several kernel functions and bandwidth estimations, in addition to the non-parametric bootstrap. The resulting data-driven methodologies can give useful probabilistic information on the software reliability prediction under the incomplete knowledge on fault count distribution.