SELGNov 26, 2020

Early Life Cycle Software Defect Prediction. Why? How?

arXiv:2011.13071v330 citations
AI Analysis

This study offers a simplification for software defect prediction, suggesting that continuous model updates may not be necessary for software developers and project managers after the initial few months.

The researchers investigated early life cycle software defect prediction and found that most defects occur very early in a project's life cycle. They showed that defect predictors learned from the first 150 commits and four months perform as well as those trained on more data.

Many researchers assume that, for software analytics, "more data is better." We write to show that, at least for learning defect predictors, this may not be true. To demonstrate this, we analyzed hundreds of popular GitHub projects. These projects ran for 84 months and contained 3,728 commits (median values). Across these projects, most of the defects occur very early in their life cycle. Hence, defect predictors learned from the first 150 commits and four months perform just as well as anything else. This means that, at least for the projects studied here, after the first few months, we need not continually update our defect prediction models. We hope these results inspire other researchers to adopt a "simplicity-first" approach to their work. Some domains require a complex and data-hungry analysis. But before assuming complexity, it is prudent to check the raw data looking for "short cuts" that can simplify the analysis.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes