The Impact of Using Regression Models to Build Defect Classifiers

TL;DR

This study compares discretized defect classifiers with regression-based classifiers across multiple datasets and models, revealing that regression approaches can outperform traditional discretized methods, especially in datasets with low defect ratios.

Contribution

The paper introduces a comparative analysis of discretized and regression-based defect classifiers, highlighting the potential advantages of regression models in defect prediction tasks.

Findings

Random forest classifiers perform best among tested models.

Discretized classifiers do not always outperform regression-based classifiers.

Regression models are particularly beneficial when defect ratios are low.

Abstract

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.