Deep Learning and Data Augmentation for Detecting Self-Admitted Technical Debt

TL;DR

This paper enhances the detection and categorization of Self-Admitted Technical Debt using deep learning models and data augmentation, addressing dataset imbalance issues to improve performance across different artifact types.

Contribution

It introduces a two-step approach with data augmentation to improve SATD detection and categorization, providing a balanced dataset and demonstrating significant performance improvements.

Findings

Data augmentation improves SATD detection accuracy.

Two-step approach effectively categorizes different SATD types.

Balanced dataset benefits future SATD research.

Abstract

Self-Admitted Technical Debt (SATD) refers to circumstances where developers use textual artifacts to explain why the existing implementation is not optimal. Past research in detecting SATD has focused on either identifying SATD (classifying SATD items as SATD or not) or categorizing SATD (labeling instances as SATD that pertain to requirement, design, code, test debt, etc.). However, the performance of these approaches remains suboptimal, particularly for specific types of SATD, such as test and requirement debt, primarily due to extremely imbalanced datasets. To address these challenges, we build on earlier research by utilizing BiLSTM architecture for the binary identification of SATD and BERT architecture for categorizing different types of SATD. Despite their effectiveness, both architectures struggle with imbalanced data. Therefore, we employ a large language model data augmentation strategy to mitigate this issue. Furthermore, we introduce a two-step approach to identify and categorize SATD across various datasets derived from different artifacts. Our contributions include providing a balanced dataset for future SATD researchers and demonstrating that our approach significantly improves SATD identification and categorization performance compared to baseline methods.