改进PSO在软件缺陷预测模型中的研究

摘要

The adaptive dynamic and median particle swarm optimization(ADPSO) algorithm is introduced, Which is used to soft-ware defect prediction models. The model in this paper extracts the relationship rules of software quality and metrics. Information entropy approach is applied to simplify the extraction rule set. The result shows that this method set of rules can be streamlined and improves the forecast accuracy.%文中介绍了运用改进PSO粒子群优化算法，并将该算法应用到软件缺陷预测模型。在该模型中提取度量属性与软件质量属性之间的关系规则，并运用互信息熵的方法简化提取规则集。与标准粒子群优化进行了比较，实验表明该方法实现精简规则集的同时又提高了预测的准确率。