English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The interpretation of defect models heavily relies on software metrics thatare used to construct them. However, such software metrics are often correlatedto defect models. Prior work often uses feature selection techniques to removecorrelated metrics in order to improve the performance of defect models. Yet,the interpretation of defect models may be misleading if feature selectiontechniques produce subsets of inconsistent and correlated metrics. In thispaper, we investigate the consistency and correlation of the subsets of metricsthat are produced by nine commonly-used feature selection techniques. Through acase study of 13 publicly-available defect datasets, we find that featureselection techniques produce inconsistent subsets of metrics and do notmitigate correlated metrics, suggesting that feature selection techniquesshould not be used and correlation analyses must be applied when the goal ismodel interpretation. Since correlation analyses often involve manual selectionof metrics by a domain expert, we introduce AutoSpearman, an automated metricselection approach based on correlation analyses. Our evaluation indicates thatAutoSpearman yields the highest consistency of subsets of metrics amongtraining samples and mitigates correlated metrics, while impacting modelperformance by 1-2%pts. Thus, to automatically mitigate correlated metrics wheninterpreting defect models, we recommend future studies use AutoSpearman inlieu of commonly-used feature selection techniques.

AutoSpearman: Automatically Mitigating Correlated Metrics for Interpreting Defect Models