Abstract Self-Admitted Technical Debt (SATD) refers to technical compromises explicitly admitted by developers in natural language artifacts, such as code comments, commit messages, and issue trackers. Among its types, Architecture Technical Debt (ATD) is particularly difficult to detect due to its abstract and context-dependent nature. Manual annotation of ATD is costly, time-consuming, and challenging to scale. To reduce labeling effort, this study combines keyword-based filtering, active learning, and explainable AI for ATD detection. We refined an existing dataset of ATD-related Jira issues to obtain an expert-validated seed set used to extract representative keywords. These keywords were then applied to identify more than 103k candidate issues across 10 open-source projects. To assess the reliability of keyword-based filtering, we qualitatively evaluated a statistically representative sample of labeled issues. Building on the resulting dataset, we applied active learning with multiple query strategies to prioritize informative samples for annotation. The results show that Breaking Ties achieved the best performance, with an F1-score of 0.72 and a 49% reduction in annotation effort. To improve transparency, we used SHAP and LIME to explain ATD classification results. Expert evaluation showed that both methods provided useful explanations, with LIME generally preferred for its clarity and ease of use.

Architectural technical debt (ATD) represents trade-offs in software architecture that accelerate initial development but create long-term maintenance challenges. ATD, in particular when self-admitted, impacts the foundational structure of software, making it difficult to detect and resolve. This study investigates the lifecycle of ATD, focusing on how it affects i) the connectivity between classes and ii) the frequency of file modifications. We aim to understand how ATD evolves from introduction to repayment and its implications on software architectures. Our empirical approach was applied to a dataset of SATD items extracted from various software artifacts. We isolated ATD instances, filtered for architectural indicators, and calculated dependencies at different lifecycle stages using FAN-IN and FAN-OUT metrics. Statistical analyses, including the Mann-Whitney U test and Cliff's Delta, were used to assess the significance and effect size of connectivity and dependency changes over time. We observed that ATD repayment increased class connectivity, with FAN-IN increasing by 57.5% on average and FAN-OUT by 26.7%, suggesting a shift toward centralization and increased architectural complexity after repayment. Moreover, ATD files were modified less frequently than Non-ATD files, with changes accumulated in high-dependency portions of the code. Our study shows that resolving ATD improves software quality in the short-term, but can make the architecture more complex by centralizing dependencies. Also, even if dependency metrics (like FAN-IN and FAN-OUT) can help understand the impact of ATD, they should be combined with other measures to capture other effects of ATD on software maintainability.

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.

Self-admitted technical debt (SATD) refers to a form of technical debt in which developers explicitly acknowledge and document the existence of technical shortcuts, workarounds, or temporary solutions within the codebase. Over recent years, researchers have manually labeled datasets derived from various software development artifacts: source code comments, messages from the issue tracker and pull request sections, and commit messages. These datasets are designed for training, evaluation, performance validation, and improvement of machine learning and deep learning models to accurately identify SATD instances. However, class imbalance poses a serious challenge across all the existing datasets, particularly when researchers are interested in categorizing the specific types of SATD. In order to address the scarcity of labeled data for SATD identification (i.e., whether an instance is SATD or not) and categorization (i.e., which type of SATD is being classified) in existing datasets, we share the SATDAUG dataset, an augmented version of existing SATD datasets, including source code comments, issue tracker, pull requests, and commit messages. These augmented datasets have been balanced in relation to the available artifacts and provide a much richer source of labeled data for training machine learning or deep learning models.