一种非参数化SRGM预测能力的研究

Research on a non-parametric SRGM prediction ability

提出了一种非参数化软件可靠性增长模型的改进,该模型考虑软件工程的多样性,利用支持向量回归技术并结合机器学习相关知识,完成故障检测和预测过程。通过三组真实的失效数据集的实验,将所改进方法与传统软件可靠性增长模型比较,结果显示,改进的非参数化软件可靠性增长模型具有更好的模型通用性和预测性能。

Software reliability evaluation performance directly affects the workload of software testing.This paper proposes an improved model of nonparametric software reliability growth model,which considers the diversity of software engineering,using support vector regression technology combined with machine learning related knowledge,complete the fault detection and prediction process.Through the experiments of three sets of real failure data sets,the improved method is compared with the traditional software reliability growth model.The results show that the improved nonparametric software reliability growth model has better model versatility and predictive performance.