arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.02389cs.LGcs.AIcs.SE
收藏

特征重要性方法对缺陷分类器解释的影响

The impact of feature importance methods on the interpretation of defect classifiers

Huawei(华为);Queen’s University(女王大学);University of Manitoba(曼尼托巴大学);Kyushu University(九州大学)

Gopi Krishnan Rajbahadur, Shaowei Wang, Yasutaka Kamei, Ahmed E. Hassan

摘要:

分类器特定(CS)和分类器不可知(CA)的特征重要性方法被广泛使用(通常可互换)由现有的研究,从缺陷分类器获得特征重要性排名。然而,即使对于相同的数据集和分类器,不同的特征重要性方法也可能计算不同的特征重要性排名。因此,这种可互换使用的特征重要性的方法可能会导致结论不稳定,除非有一个很强的协议之间的不同的方法。因此,在本文中,我们通过对18个软件项目和6个常用分类器的案例研究,评估与所研究的分类器相关的特征重要性等级之间的协议。我们发现:1)CA和CS方法计算的特征重要性排名并不总是完全一致。2)所研究的CA方法计算的特征重要性排名表现出很强的一致性,包括在给定的数据集和分类器的前1名和前3名的排名报告的功能,而即使是常用的CS方法产生截然不同的特征重要性排名。这些发现引起了人们对重复研究结论稳定性的担忧。我们进一步观察到,常用的缺陷数据集充满了特征交互,这些特征交互影响CS方法(而不是CA方法)的计算特征重要性排名。我们表明,删除这些功能的相互作用,即使是简单的方法,如CFS提高CA和CS方法的计算特征的重要性等级之间的协议。根据我们的研究结果,我们为利益相关者和从业者提供了进行模型解释的指导方针,并为未来的研究提供了方向,例如,未来的研究需要调查高级特征交互去除方法对不同CS方法的计算特征重要性等级的影响。

英文摘要:

Classifier specific (CS) and classifier agnostic (CA) feature importance methods are widely used (often interchangeably) by prior studies to derive feature importance ranks from a defect classifier. However, different feature importance methods are likely to compute different feature importance ranks even for the same dataset and classifier. Hence such interchangeable use of feature importance methods can lead to conclusion instabilities unless there is a strong agreement among different methods. Therefore, in this paper, we evaluate the agreement between the feature importance ranks associated with the studied classifiers through a case study of 18 software projects and six commonly used classifiers. We find that: 1) The computed feature importance ranks by CA and CS methods do not always strongly agree with each other. 2) The computed feature importance ranks by the studied CA methods exhibit a strong agreement including the features reported at top-1 and top-3 ranks for a given dataset and classifier, while even the commonly used CS methods yield vastly different feature importance ranks. Such findings raise concerns about the stability of conclusions across replicated studies. We further observe that the commonly used defect datasets are rife with feature interactions and these feature interactions impact the computed feature importance ranks of the CS methods (not the CA methods). We demonstrate that removing these feature interactions, even with simple methods like CFS improves agreement between the computed feature importance ranks of CA and CS methods. In light of our findings, we provide guidelines for stakeholders and practitioners when performing model interpretation and directions for future research, e.g., future research is needed to investigate the impact of advanced feature interaction removal methods on computed feature importance ranks of different CS methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑