An Adaptive Exponential Smoothing Approach for Software Reliability Prediction

In this paper, we propose an adaptive exponential smoothing approach for software reliability prediction. We first explain the exponential smoothing technique from the mathematical viewpoints of software reliability modeling. That is, we will show how to apply adaptive exponential smoothing to predict the number of software failures that will occur next time by treating software failure data as time series. Furthermore, we will use this technique to build an adaptive exponential smoothing software reliability model. The results of comparisons with other models on actual data set are presented. From the results, we can see that our model is better than the traditional G-O model in these experiments.