论文部分内容阅读
为提高航天测控软件的质量与可靠性,提出一种基于改进的PSO-SVM(Particle Swarm Optimization-Support Vector Machine,粒子群优化支持向量机)方法的航天测控软件缺陷预测模型。针对航天测控软件领域特征,构造了基于软件生命周期的软件度量集,并收集了实际航天测控软件的度量和缺陷数据,通过对软件历史版本数据的学习,在软件当前版本的生命周期早期数据的基础上进行缺陷预测。实例应用结果表明,采用历史版本软件数据对当前软件版本进行缺陷预测,从全局来看可达90%的预测准确度。因此,该方法可用于对航天测控软件的缺陷预测。
In order to improve the quality and reliability of aerospace measurement and control software, a defect prediction model of aerospace measurement and control software based on the improved PSO-SVM (Particle Swarm Optimization-Support Vector Machine) method is proposed. Aiming at the characteristics of aerospace measurement and control software, a software measurement set based on the software life cycle is constructed and the measurement and defect data of the actual aerospace measurement and control software are collected. Through the learning of the historical version of the software and the data of the early life of the software Based on the defect prediction. The application of the example shows that using the historical version of the software data to predict the defect of the current software version, the prediction accuracy of 90% can be achieved globally. Therefore, this method can be used to predict defects in aerospace measurement and control software.