AutoSpearman: Automatically Mitigating Correlated Metrics for Interpreting Defect Models
This addresses the issue for software engineering researchers and practitioners using defect models, but it is incremental as it builds on existing correlation analyses by automating them.
The paper tackles the problem of misleading interpretations in defect models due to inconsistent and correlated metrics from feature selection techniques, finding that these techniques produce inconsistent subsets and fail to mitigate correlations. It introduces AutoSpearman, an automated approach based on correlation analyses, which yields the highest consistency and mitigates correlated metrics while impacting model performance by 1-2% points.
The interpretation of defect models heavily relies on software metrics that are used to construct them. However, such software metrics are often correlated to defect models. Prior work often uses feature selection techniques to remove correlated metrics in order to improve the performance of defect models. Yet, the interpretation of defect models may be misleading if feature selection techniques produce subsets of inconsistent and correlated metrics. In this paper, we investigate the consistency and correlation of the subsets of metrics that are produced by nine commonly-used feature selection techniques. Through a case study of 13 publicly-available defect datasets, we find that feature selection techniques produce inconsistent subsets of metrics and do not mitigate correlated metrics, suggesting that feature selection techniques should not be used and correlation analyses must be applied when the goal is model interpretation. Since correlation analyses often involve manual selection of metrics by a domain expert, we introduce AutoSpearman, an automated metric selection approach based on correlation analyses. Our evaluation indicates that AutoSpearman yields the highest consistency of subsets of metrics among training samples and mitigates correlated metrics, while impacting model performance by 1-2%pts. Thus, to automatically mitigate correlated metrics when interpreting defect models, we recommend future studies use AutoSpearman in lieu of commonly-used feature selection techniques.