The Impact of Using Regression Models to Build Defect Classifiers
The Impact of Using Regression Models to Build Defect Classifiers
复制标题
DOI：
10.1109/msr.2017.4
复制
发表时间：
2017-05
期刊：
影响因子：
--
通讯作者：
Gopi Krishnan Rajbahadur;Shaowei Wang;Yasutaka Kamei;A. Hassan
中科院分区：
文献类型：
--
作者：
Gopi Krishnan Rajbahadur;Shaowei Wang;Yasutaka Kamei;A. Hassan
通常的做法是将连续缺陷计数离散化为缺陷和非缺陷类，并在构建缺陷分类器（离散化分类器）时将它们用作目标变量。然而，这种连续缺陷计数的离散化导致信息丢失，这可能会影响缺陷分类器的性能和解释。构建缺陷分类器的另一种可能的方法是通过使用回归模型，然后将预测的缺陷计数离散化为有缺陷和无缺陷的类别（基于回归的分类器）。在本文中，我们比较了使用这两种方法（即，离散化分类器和基于回归的分类器）跨越六个常用的机器学习分类器（即，线性/逻辑回归，随机森林，KNN，SVM，CART和神经网络）和17个数据集。我们发现：i）对于两种分类器构建方法，基于随机森林的分类器优于其他分类器（最佳AUC）。离散化分类器）并不总是导致更好的性能。因此，我们建议未来的缺陷分类研究应该考虑建立基于回归的分类器（特别是当模型数据集的缺陷率较低时）。此外，我们建议，这两种方法建立缺陷分类器应进行探索，所以最好的性能分类器时，可以使用确定最有影响力的功能。
It is common practice to discretize continuous defect counts into defective and non-defective classes and use them as a target variable when building defect classifiers (discretized classifiers). However, this discretization of continuous defect counts leads to information loss that might affect the performance and interpretation of defect classifiers. Another possible approach to build defect classifiers is through the use of regression models then discretizing the predicted defect counts into defective and non-defective classes (regression-based classifiers). In this paper, we compare the performance and interpretation of defect classifiers that are built using both approaches (i.e., discretized classifiers and regression-based classifiers) across six commonly used machine learning classifiers (i.e., linear/logistic regression, random forest, KNN, SVM, CART, and neural networks) and 17 datasets. We find that: i) Random forest based classifiers outperform other classifiers (best AUC) for both classifier building approaches, ii) In contrast to common practice, building a defect classifier using discretized defect counts (i.e., discretized classifiers) does not always lead to better performance. Hence we suggest that future defect classification studies should consider building regression-based classifiers (in particular when the defective ratio of the modeled dataset is low). Moreover, we suggest that both approaches for building defect classifiers should be explored, so the best-performing classifier can be used when determining the most influential features.