基于投影寻踪的Web软件复杂性度量  

Web software complexity metrics based on projection pursuit

作　　者：曾一[1] 胡小威[1] 李鹃[1] 

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《计算机应用》2012年第3期827-830,共4页journal of Computer Applications

摘　　要：传统的软件复杂性度量方法主要是针对C/C++、Ada等语言开发的非Web应用。以面向对象的基于Struts框架的Web软件为研究对象,提出了适合于Web-Struts软件的3个方面的复杂性度量指标,并提出了一种基于带交叉算子人工鱼群和投影寻踪(PP)算法的Web应用软件复杂性度量方法。把Web软件多个复杂性度量指标综合成一维综合投影值,利用样本数据求解最佳投影方向,确定评价等级的综合投影值区间,根据测试样本综合投影值与区间值比较,获得综合评价结果。实例评价结果表明,所提方法具有较强的适用性和应用性。Web software complexity metrics does play a very important role in the software development.The traditional software complexity metrics method mainly targets on the non-Web applications which use language like C/C ＋ ＋ and Ada.This paper took object-oriented Web software based on Struts framework as research subject and put forward three complexity metrics suitable for the Web-Struts software.Besides,this paper also proposed a method for computing Web software complexity metrics based on Artificial Fish Swarm Algorithm（AFSA） with cross operator and Projection Pursuit（PP） algorithm.After integrating multiple complexity metrics into one-dimension comprehensive projection value,the optimized projection direction could be acquired through sample data.Then the comprehensive projection value of evaluation grades could also be determined.According to the comparison between the comprehensive projection values of the testing samples and the interval of level,the comprehensive metrics result could be finally obtained.The example evaluation results prove the feasibility and effectiveness of the proposed method.

关 键 词：软件复杂性度量 面向对象 STRUTS框架 投影寻踪 交叉算子 人工鱼群算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]