基于加权估计的软件实验室能力比对测试结果评价  

The Weight and Evaluation Model for Software Testing Ability Comparison among Software Testing Laboratories

作　　者：王勇利[1] 王艳军[1] 张海军[1] 

机构地区：[1]中国人民解放军91404部队

出　　处：《实验室研究与探索》2015年第3期246-250,共5页Research and Exploration In Laboratory

摘　　要：为实现对软件测评实验室能力比对测试结果的定量分析与评价,本文在融合参加能力比对测试各方所发现软件缺陷的基础上,以测试广度和测试深度作为评价要素,选取正态分布作为趋势估计曲线,通过调整诸权重及系数、归一化趋势估计曲线幅度等手段,统一评价基准,构建出一种能力比对测试结果评价体系模型。还对此模型组织实施、关键过程实现手段等方面进行详细阐述。该评价模型可直接用于对软件测评实验室能力比对测试结果的定量分析与评价,对于构建新的定量评价体系也具有较强的借鉴意义。In the testing ability comparative test for software testing laboratories, it is hard to estimate the results of every laboratory quantitatively. To solve the intractable conundrum, this paper advances the ＂Weight and Evaluation Model＂ （the WE model） , an evaluating model for software testing ability comparison, based on weighting evaluation, curve fitting and trending. The WE model takes the extent and depth of every software testing activities synthetically and synchronously, and can evaluate each lab＇s testing results quantitatively according to the same criteria, This paper also indicates the crucial actualizing means and method for organizing and actualizing the WE model.

关 键 词：软件测评 能力比对 定量评价 等价类划分 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]