【摘要】 开源软件缺陷问题是目前软件工程领域一个非常重要的研究领域，为了避免由于软件缺陷而引发的故障，如何识别和预测软件系统中存在的缺陷也是目前一个重要研究课题.针对上述现状和问题，提出了量子粒子群混合烟花优化算法，用于对支持向量机进行参数寻优，将两种算法进行并联式融合，从而达到更好的寻优效果.其次，为验证提出方法的有效性，在NASA MDP数据集上进行仿真实验，将量子粒子群算法、混合烟花算法的量子粒子群算法用于SVM的参数寻优，并且与经典的SVM之间进行对比，证明其有效性.仿真实验结果表明，提出的基于量子粒子群混合烟花算法优化支持向量机软件缺陷预测模型的综合性能要优于其他模型，提出的量子粒子群混合烟花算法对量子粒子群跳出局部最优的能力有较大程度的提高.更多还原

【Abstract】 The problem of open source software defects is a very important research field in the field of software engineering at present.In order to avoid the faults caused by software defects, how to identify and predict the defects in software systems is also an important research topic at present.In view of the above situation and problems, this paper proposed a hybrid optimization algorithm of QPSO and FWA,which was used to optimize the parameters of support vector machine, and the two algorithms were merged in parallel to achieve better optimization results.In order to verify the effectiveness of the proposed method, a simulation experiment was carried out on NASA MDP data set, QPSO and a hybrid optimization algorithm of QPSO and FWA were used to optimize the parameters of SVM,and the comparison with the classical SVM proved its effectiveness.The simulation results showed that the comprehensive performance of the software defect prediction model based on SVM optimized by a hybrid algorithm of QPSO and FWA was better than other models, and the proposed algorithm could greatly improve the ability of QPSO to jump out of local optimum.更多还原