pith. sign in

arxiv: 2312.15020 · v4 · submitted 2023-12-19 · 💻 cs.SE · cs.AI

Self-Admitted Technical Debt Detection Approaches: A Decade Systematic Review

Edi Sutoyo , Andrea Capiluppi This is my paper

Pith reviewed 2026-05-24 04:56 UTC · model grok-4.3

classification 💻 cs.SE cs.AI
keywords self-admitted technical debtSATD detectionsystematic literature reviewmachine learningdeep learningtransformer modelssoftware engineeringcode comments
0
0 comments X

The pith

A decade-long review finds that deep learning and transformer models have raised accuracy in detecting self-admitted technical debt beyond early heuristic methods, though scaling to industry use stays difficult.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper reviews research on finding self-admitted technical debt, where developers note their own shortcuts in code comments. It follows the shift from basic rule-based searches starting in 2014 to machine learning, deep learning, and transformer approaches by 2025. The analysis compares how well each type performs using standard measures such as precision, recall, and F1 score. It concludes that later models detect more of these debts correctly, yet problems with varied data sets and model reuse limit wider use. This matters for software teams because better detection could reduce hidden future costs from rushed decisions.

Core claim

The review establishes that heuristic-based techniques formed the starting point for SATD detection, while later deep learning and transformer models have delivered measurable gains in detection accuracy as shown by higher precision, recall, and F1 scores across the examined studies; however, dataset differences, limited generalizability, and lack of explainability continue to block broader industrial application.

What carries the argument

The systematic classification and metric-based comparison of SATD detection methods across papers from 2014 to early 2025, grouped by technique type from heuristics through ML, DL, and transformers.

If this is right

  • Transformer-based detectors achieve higher F1 scores than the early comment-matching rules.
  • Heterogeneous data sets prevent reliable ranking of one approach over another.
  • Explainability remains low in the newest models, limiting trust in their outputs.
  • Practical tools for developers will need work on model reuse across projects and languages.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same review structure could be applied to other forms of technical debt that are not self-admitted.
  • Combining SATD detectors with automated refactoring suggestions might shorten the time from detection to fix.
  • Studies that test the same models on non-English code comments would check whether language-specific patterns affect the reported gains.

Load-bearing premise

The review assumes that performance numbers reported on different data sets and test setups can be compared directly and that the chosen papers represent the field without major gaps.

What would settle it

A single large-scale experiment that applies every reviewed method to one shared, balanced data set and finds no accuracy improvement from deep learning or transformer models over the original heuristics would undermine the central finding.

read the original abstract

Technical debt (TD) refers to the long-term costs associated with suboptimal design or code decisions in software development, often made to meet short-term delivery goals. Self-Admitted Technical Debt (SATD) occurs when developers explicitly acknowledge these trade-offs in the codebase, typically through comments or annotations. SATD detection has become an increasingly important research area, particularly with the rise of learning-based techniques that aim to streamline SATD detection. This systematic literature review provides a comprehensive analysis of SATD detection approaches published between 2014 and early 2025, focusing on the evolution of techniques from heuristic-based techniques to more advanced ML, DL, and Transformer-based models. It examines key trends in SATD detection methodologies and tools, evaluates the effectiveness of different approaches using metrics like precision, recall, and F1 score, and highlights the primary challenges in this domain, including dataset heterogeneity, model generalizability, and explainability. The findings reveal that while early heuristic-based techniques laid the foundation for SATD detection, more recent advancements in DL and Transformer models have significantly improved detection accuracy. However, challenges remain in scaling these models for broader industrial adoption. This review offers insights into current research gaps and provides directions for future work, aiming to improve the robustness and practicality of SATD detection tools.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. This systematic literature review examines self-admitted technical debt (SATD) detection approaches published between 2014 and early 2025. It traces the evolution from heuristic-based techniques to machine learning, deep learning, and Transformer models, aggregates reported precision/recall/F1 scores to evaluate effectiveness, and identifies challenges including dataset heterogeneity, model generalizability, and explainability. The central claim is that recent DL/Transformer advances have significantly improved detection accuracy over earlier methods, while scaling for industrial use remains difficult.

Significance. If the comparative claims can be substantiated through adjusted cross-study analysis, the review would provide a useful decade-long synthesis of SATD detection research and surface actionable gaps. The explicit focus on challenges such as heterogeneous datasets is a strength, but the absence of quantitative synthesis or normalization currently limits the paper's ability to support strong trend conclusions.

major comments (2)
  1. [Results section] Results section (and abstract): The claim that DL and Transformer models have 'significantly improved detection accuracy' rests on narrative aggregation of per-paper F1/precision values. Because the underlying studies employ distinct codebases, comment distributions, train/test splits, and labeling schemes, raw metric values are not commensurable; no dataset-size normalization, cross-study statistical tests, or common-benchmark re-evaluation is reported. This directly undermines the central comparative finding.
  2. [Methodology section] Methodology section: The review does not supply the search strings, database list, exact inclusion/exclusion criteria, or quality-assessment protocol used to select papers. Without these details, it is impossible to evaluate completeness, replicability, or the risk of publication bias in the sampled literature.
minor comments (1)
  1. [Abstract] The abstract states coverage through 'early 2025' while the arXiv identifier indicates a 2023 submission; clarify the actual search cutoff date and any update process in the introduction or methodology.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Results section] Results section (and abstract): The claim that DL and Transformer models have 'significantly improved detection accuracy' rests on narrative aggregation of per-paper F1/precision values. Because the underlying studies employ distinct codebases, comment distributions, train/test splits, and labeling schemes, raw metric values are not commensurable; no dataset-size normalization, cross-study statistical tests, or common-benchmark re-evaluation is reported. This directly undermines the central comparative finding.

    Authors: We agree that raw metrics from heterogeneous studies are not directly commensurable and that the absence of normalization or statistical cross-study analysis limits the strength of comparative claims. Our review reports the metrics as published to document trends in the literature rather than asserting strict quantitative superiority. In revision we will qualify all comparative statements in the abstract and results, add an explicit discussion of comparability limitations, and include a threats-to-validity subsection addressing dataset and experimental heterogeneity. A full quantitative meta-analysis or re-evaluation on common benchmarks is not feasible within the scope of this review given the diversity of primary studies. revision: partial

  2. Referee: [Methodology section] Methodology section: The review does not supply the search strings, database list, exact inclusion/exclusion criteria, or quality-assessment protocol used to select papers. Without these details, it is impossible to evaluate completeness, replicability, or the risk of publication bias in the sampled literature.

    Authors: We acknowledge that these protocol details were omitted. The revised manuscript will include the complete search strings, the list of databases queried, the precise inclusion/exclusion criteria, and the quality-assessment protocol. These additions will follow established systematic-review reporting standards and enable assessment of replicability and bias risk. revision: yes

Circularity Check

0 steps flagged

No circularity: literature review with no derivations or fitted predictions

full rationale

This is a systematic literature review paper that aggregates and narrates findings from prior publications on SATD detection. It contains no original equations, derivations, parameter fits, predictions, or uniqueness theorems. The central claim of improved accuracy in recent DL/Transformer models is presented as a summary of reported metrics across studies rather than a self-derived result. No steps match any of the enumerated circularity patterns (self-definitional, fitted-input-called-prediction, self-citation load-bearing, etc.), and the paper is self-contained as a descriptive survey.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a systematic literature review that summarizes existing research without introducing new parameters, axioms, or entities. The central claims rest on the completeness and comparability of prior publications.

pith-pipeline@v0.9.0 · 5757 in / 1053 out tokens · 19903 ms · 2026-05-24T04:56:22.210489+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

109 extracted references · 109 canonical work pages

  1. [1]

    A tertiary study on technical debt: Types, management strategies, research trends, and base information for practitioners.Information and Software Technology.2018;102:117–145

    Rios N, Mendonça Neto dMG, Spínola RO. A tertiary study on technical debt: Types, management strategies, research trends, and base information for practitioners.Information and Software Technology.2018;102:117–145. doi: 10.1016/j.infsof.2018.05.010

    work page doi:10.1016/j.infsof.2018.05.010 2018

  2. [2]

    An enterprise perspective on technical debt.Proceedings - International Conference on Software Engineering.2011:35–38

    Klinger T, Tarr P, Wagstrom P, Williams C. An enterprise perspective on technical debt.Proceedings - International Conference on Software Engineering.2011:35–38. doi: 10.1145/1985362.1985371

    work page doi:10.1145/1985362.1985371 2011

  3. [3]

    Embracing technical debt, from a startup company perspective

    Besker T, Martini A, Edirisooriya Lokuge R, Blincoe K, Bosch J. Embracing technical debt, from a startup company perspective. Proceedings - 2018 IEEE International Conference on Software Maintenance and Evolution, ICSME 2018.2018:415–425. doi: 10.1109/ICSME.2018.00051

    work page doi:10.1109/icsme.2018.00051 2018

  4. [4]

    Managing Technical Debt in Software Engineering.Dagstuhl Reports.2016;6(4):110–

    Avgeriou P, Kruchten P, Ozkaya I, Seaman C. Managing Technical Debt in Software Engineering.Dagstuhl Reports.2016;6(4):110–

    work page 2016

  5. [5]

    doi: 10.4230/DagRep.6.4.110

    work page doi:10.4230/dagrep.6.4.110

  6. [6]

    The WyCash portfolio management system.Proceedings of the Conference on Object-Oriented Programming Systems, Languages, and Applications, OOPSLA.1992;Part F1296(October):29–30

    Cunningham W. The WyCash portfolio management system.Proceedings of the Conference on Object-Oriented Programming Systems, Languages, and Applications, OOPSLA.1992;Part F1296(October):29–30. doi: 10.1145/157709.157715

    work page doi:10.1145/157709.157715 1992

  7. [7]

    Managing technical debt in software-reliant systems

    Brown N, Cai Y, Guo Y, et al. Managing technical debt in software-reliant systems. In:Association for Computing Machinery. 2010:47–51

    work page 2010

  8. [8]

    An exploration of technical debt.Journal of Systems and Software.2013;86(6):1498–1516

    Tom E, Aurum A, Vidgen R. An exploration of technical debt.Journal of Systems and Software.2013;86(6):1498–1516. doi: 10.1016/j.jss.2012.12.052

    work page doi:10.1016/j.jss.2012.12.052 2013

  9. [9]

    An exploratory study on self-admitted technical debt

    Potdar A, Shihab E. An exploratory study on self-admitted technical debt. In:Institute of Electrical and Electronics Engineers. 2014:91–100

    work page 2014

  10. [10]

    A survey of self-admitted technical debt.Journal of Systems and Software.2019;152:70–82

    Sierra G, Shihab E, Kamei Y. A survey of self-admitted technical debt.Journal of Systems and Software.2019;152:70–82. doi: 10.1016/j.jss.2019.02.056

    work page doi:10.1016/j.jss.2019.02.056 2019

  11. [11]

    Guidelines for performing Systematic Literature Reviews in Software Engineering

    Kitchenham B, Charters SM. Guidelines for performing Systematic Literature Reviews in Software Engineering. tech. rep., Keele University; Keele, Newcastle ST5 5BG, UK: 2007

    work page 2007

  12. [12]

    Measuring and monitoring technical debt

    Seaman C, Guo Y. Measuring and monitoring technical debt. In: , , . 82. Elsevier, 2011:25–46

    work page 2011

  13. [13]

    An exploration of technical debt.Journal of Systems and Software.2013;86(6):1498–1516

    Tom E, Aurum A, Vidgen R. An exploration of technical debt.Journal of Systems and Software.2013;86(6):1498–1516

    work page 2013

  14. [14]

    Technical debt: From metaphor to theory and practice.IEEE Software.2012;29(6):18–21

    Kruchten P, Nord RL, Ozkaya I. Technical debt: From metaphor to theory and practice.IEEE Software.2012;29(6):18–21. doi: 10.1109/MS.2012.167 SELF-ADMITTED TECHNICAL DEBT DETECTION APPROACHES: A DECADE SYSTEMATIC REVIEW 25

    work page doi:10.1109/ms.2012.167 2012

  15. [15]

    A case study on effectively identifying technical debt

    Zazworka N, Spínola RO, Vetro A, Shull F, Seaman C. A case study on effectively identifying technical debt. In:Association for Computing Machinery. 2013:42–47

    work page 2013

  16. [16]

    Investigating technical debt folklore: Shedding some light on tech- nical debt opinion.2013 4th International Workshop on Managing Technical Debt, MTD 2013 - Proceedings.2013:1–7

    Spínola RO, Vetrò A, Zazworka N, Seaman C, Shull F. Investigating technical debt folklore: Shedding some light on tech- nical debt opinion.2013 4th International Workshop on Managing Technical Debt, MTD 2013 - Proceedings.2013:1–7. doi: 10.1109/MTD.2013.6608671

    work page doi:10.1109/mtd.2013.6608671 2013

  17. [17]

    Liddy ED.Natural language processing. NY. Marcel Decker, Inc. 2nd edition ed., 2001

    work page 2001

  18. [18]

    Natural language processing.Annual Review of Information Science and Technology.2003;37:51–89

    Chowdhury GG. Natural language processing.Annual Review of Information Science and Technology.2003;37:51–89. doi: 10.1002/ARIS.1440370103

    work page doi:10.1002/aris.1440370103 2003

  19. [19]

    Chapman and Hall/CRC, 2010

    Indurkhya N, Damerau FJ, Indurkhya N, Damerau FJ.Handbook of natural language processing. Chapman and Hall/CRC, 2010

    work page 2010

  20. [20]

    Natural Language Processing

    Chowdhary KR. Natural Language Processing. In: , , Springer, 2020:603–649

    work page 2020

  21. [21]

    The use of text retrieval and natural language processing in software engineering

    Haiduc S, Arnaoudova V, Marcus A, Antoniol G. The use of text retrieval and natural language processing in software engineering. In: Institute of Electrical and Electronics Engineers . 2016:898–899

    work page 2016

  22. [22]

    Automatic Identification of Self-Admitted Technical Debt from Different Sources.arXiv preprint arXiv:2202.02387

    Li Y, Soliman M, Avgeriou P. Automatic Identification of Self-Admitted Technical Debt from Different Sources.arXiv preprint arXiv:2202.02387. 2022. doi: 10.48550/arXiv.2202.02387

    work page doi:10.48550/arxiv.2202.02387 2022

  23. [23]

    Todo or to bug: exploring how task annotations play a role in the work practices of software developers

    Storey MA, Ryall J, Bull RI, Myers D, Singer J. Todo or to bug: exploring how task annotations play a role in the work practices of software developers. In:Institute of Electrical and Electronics Engineers . 2008:251–260

    work page 2008

  24. [24]

    Which documentation for software maintenance?.Journal of the Brazilian Computer Society

    Souza dSCB, Anquetil N, Oliveira dKM. Which documentation for software maintenance?.Journal of the Brazilian Computer Society. 2006;12:31–44. doi: 10.1007/BF03194494

    work page doi:10.1007/bf03194494 2006

  25. [25]

    Can development work describe itself?

    Maalej W, Happel HJ. Can development work describe itself?. In: IEEE. 2010:191–200

    work page 2010

  26. [26]

    Why so complicated? simple term filtering and weighting for location-based bug report assignment recommendation

    Shokripour R, Anvik J, Kasirun ZM, Zamani S. Why so complicated? simple term filtering and weighting for location-based bug report assignment recommendation. In: IEEE. 2013:2–11

    work page 2013

  27. [27]

    Procedures for Performing Systematic Reviews.Keele University Technical Report.2004;33(2004):1–26

    Barbara Kitchenham . Procedures for Performing Systematic Reviews.Keele University Technical Report.2004;33(2004):1–26

    work page 2004

  28. [28]

    Evidence-based software engineering and systematic literature reviews

    Kitchenham B. Evidence-based software engineering and systematic literature reviews. In: . 4034 LNCS. Springer. Springer Verlag 2006:3

    work page 2006

  29. [29]

    John Wiley \& Sons, 2008

    Petticrew M, Roberts H.Systematic reviews in the social sciences: A practical guide. John Wiley \& Sons, 2008

    work page 2008

  30. [30]

    Developing guidance on the conduct of narrative synthesis in systematic reviews.J Epidemiol Community Health.2005;59(Suppl 1):A7

    Popay J, Roberts H, Sowden A, et al. Developing guidance on the conduct of narrative synthesis in systematic reviews.J Epidemiol Community Health.2005;59(Suppl 1):A7

    work page 2005

  31. [31]

    Guidelines for snowballing in systematic literature studies and a replication in software engineering

    Wohlin C. Guidelines for snowballing in systematic literature studies and a replication in software engineering. In:Association for Computing Machinery. 2014:1–10

    work page 2014

  32. [32]

    On the performance of hybrid search strategies for systematic literature reviews in software engineering.Information and software technology.2020;123:106294

    Mourão E, Pimentel JF, Murta L, Kalinowski M, Mendes E, Wohlin C. On the performance of hybrid search strategies for systematic literature reviews in software engineering.Information and software technology.2020;123:106294

    work page 2020

  33. [33]

    Software smell detection techniques: A systematic literature review.Journal of Software: Evolution and Process.2021;33(3):e2320

    AbuHassan A, Alshayeb M, Ghouti L. Software smell detection techniques: A systematic literature review.Journal of Software: Evolution and Process.2021;33(3):e2320

    work page 2021

  34. [34]

    Three decades of software reference architectures: A systematic mapping study

    Garcés L, Martínez-Fernández S, Oliveira L, et al. Three decades of software reference architectures: A systematic mapping study. Journal of Systems and Software.2021;179:111004

    work page 2021

  35. [35]

    Identification and measurement of Requirements Technical Debt in software development: A systematic literature review.Journal of Systems and Software.2022;194:111483

    Melo A, Fagundes R, Lenarduzzi V, Santos WB. Identification and measurement of Requirements Technical Debt in software development: A systematic literature review.Journal of Systems and Software.2022;194:111483

    work page 2022

  36. [36]

    Constructing a search strategy and searching for evidence.Am J Nurs

    Aromataris E, Riitano D. Constructing a search strategy and searching for evidence.Am J Nurs. 2014;114(5):49–56. doi: 10.1097/01.NAJ.0000446779.99522.f6

    work page doi:10.1097/01.naj.0000446779.99522.f6 2014

  37. [37]

    A Contextualized Vocabulary Model for identifying technical debt on code comments

    Farias MADF, Neto MGDM, Silva ABD, Spinola RO. A Contextualized Vocabulary Model for identifying technical debt on code comments. In: Institute of Electrical and Electronics Engineers . IEEE 2015:25–32

    work page 2015

  38. [38]

    Detecting and quantifying different types of self-admitted technical Debt.2015 IEEE 7th International Workshop on Managing Technical Debt, MTD 2015 - Proceedings.2015:9–15

    Maldonado EDS, Shihab E. Detecting and quantifying different types of self-admitted technical Debt.2015 IEEE 7th International Workshop on Managing Technical Debt, MTD 2015 - Proceedings.2015:9–15. doi: 10.1109/MTD.2015.7332619

    work page doi:10.1109/mtd.2015.7332619 2015

  39. [39]

    A large-scale empirical study on self-admitted technical debt.Proceedings - 13th Working Conference on Mining Software Repositories, MSR 2016.2016:315–326

    Bavota G, Russo B. A large-scale empirical study on self-admitted technical debt.Proceedings - 13th Working Conference on Mining Software Repositories, MSR 2016.2016:315–326. doi: 10.1145/2901739.2901742

    work page doi:10.1145/2901739.2901742 2016

  40. [40]

    Identification and Remediation of Self-Admitted Technical Debt in Issue Trackers

    Li Y, Soliman M, Avgeriou P. Identification and Remediation of Self-Admitted Technical Debt in Issue Trackers. In:Institute of Electrical and Electronics Engineers . 2020:495–503

    work page 2020

  41. [41]

    Identifying self-admitted technical debt through code comment analysis with a contextualized vocabulary.Information and Software Technology.2020;121(January 2019)

    Farias MAdF, Neto MGdM, Kalinowski M, Spínola RO. Identifying self-admitted technical debt through code comment analysis with a contextualized vocabulary.Information and Software Technology.2020;121(January 2019). doi: 10.1016/j.infsof.2020.106270

    work page doi:10.1016/j.infsof.2020.106270 2020

  42. [42]

    Towards a comprehensive self-admitted technical debt extraction technique from source code comments

    Al-Slais YS. Towards a comprehensive self-admitted technical debt extraction technique from source code comments. In:Institute of Electrical and Electronics Engineers . 2021:109–114

    work page 2021

  43. [43]

    Wait for it: identifying "On-Hold" self-admitted technical debt.Empirical Software Engineering

    Maipradit R, Treude C, Hata H, Matsumoto K. Wait for it: identifying "On-Hold" self-admitted technical debt.Empirical Software Engineering. 2020;25(5):3770–3798. doi: 10.1007/s10664-020-09854-3

    work page doi:10.1007/s10664-020-09854-3 2020

  44. [44]

    Classification of Technical Debts in Software Development Using Text Analytics

    Rajalakshmi V, Sendhilkumar S, Mahalakshmi GS. Classification of Technical Debts in Software Development Using Text Analytics. EAI/Springer Innovations in Communication and Computing.2021:313–323. doi: 10.1007/978-3-030-49795-8_31

    work page doi:10.1007/978-3-030-49795-8_31 2021

  45. [45]

    How Far Have We Progressed in Identifying Self-admitted Technical Debts? A Comprehensive Empirical Study

    Guo Z, Liu S, Liu J, et al. How Far Have We Progressed in Identifying Self-admitted Technical Debts? A Comprehensive Empirical Study. ACM Transactions on Software Engineering and Methodology.2021;30(4). doi: 10.1145/3447247

    work page doi:10.1145/3447247 2021

  46. [46]

    An empirical study on self-admitted technical debt in Dockerfile.Empirical Software Engineering.2022;27(2):49

    Azuma H, Matsumoto S, Kamei Y, Kusumoto S. An empirical study on self-admitted technical debt in Dockerfile.Empirical Software Engineering.2022;27(2):49. doi: 10.1016/j.infsof.2022.106855

    work page doi:10.1016/j.infsof.2022.106855 2022

  47. [47]

    WeakSATD: Detecting Weak Self-admitted Technical Debt.Proceedings - 2022 Mining Software Repositories Conference, MSR 2022.2022:448–453

    Russo B, Camilli M, Mock M. WeakSATD: Detecting Weak Self-admitted Technical Debt.Proceedings - 2022 Mining Software Repositories Conference, MSR 2022.2022:448–453. doi: 10.1145/3524842.3528469

    work page doi:10.1145/3524842.3528469 2022

  48. [48]

    Automated Self-Admitted Technical Debt Tracking at Commit-Level: A Language-independent Approach

    Sheikhaei MS, Tian Y. Automated Self-Admitted Technical Debt Tracking at Commit-Level: A Language-independent Approach. In: IEEE. 2023:22–26

    work page 2023

  49. [49]

    On Comment Patterns that are Good Indicators of the Presence of Self-Admitted Technical Debt and those that Lead to False Positive Items

    Farias M, Mendes TS, Mendonça MG, Spínola RO. On Comment Patterns that are Good Indicators of the Presence of Self-Admitted Technical Debt and those that Lead to False Positive Items.. In: Association for Information Systems (AIS). 2021

    work page 2021

  50. [50]

    Do we need to pay technical debt in blockchain software systems?.Connection Science.2022;34(1):2026– 2047

    Qu Y, Bao T, Chen X, et al. Do we need to pay technical debt in blockchain software systems?.Connection Science.2022;34(1):2026– 2047. 26 SUTOYO and CAPILUPPI

    work page 2022

  51. [51]

    Detecting technical debt through issue trackers.CEUR Workshop Proceedings.2017;2017(QuASoQ):59–65

    Dai K, Kruchten P. Detecting technical debt through issue trackers.CEUR Workshop Proceedings.2017;2017(QuASoQ):59–65

    work page 2017

  52. [52]

    Using Natural Language Processing to Automatically Detect Self-Admitted Technical Debt

    Maldonado EDS, Shihab E, Tsantalis N. Using Natural Language Processing to Automatically Detect Self-Admitted Technical Debt. IEEE Transactions on Software Engineering.2017;43(11):1044–1062. doi: 10.1109/TSE.2017.2654244

    work page doi:10.1109/tse.2017.2654244 2017

  53. [53]

    Enhanced feature selection using word embeddings for self-admitted technical debt identification

    Flisar J, Podgorelec V. Enhanced feature selection using word embeddings for self-admitted technical debt identification. Proceedings - 44th Euromicro Conference on Software Engineering and Advanced Applications, SEAA 2018.2018:230–233. doi: 10.1109/SEAA.2018.00045

    work page doi:10.1109/seaa.2018.00045 2018

  54. [54]

    Identifying design and requirement self-admitted technical debt using N-gram IDF

    Wattanakriengkrai S, Maipradit R, Hata H, Choetkiertikul M, Sunetnanta T, Matsumoto K. Identifying design and requirement self-admitted technical debt using N-gram IDF. In:Institute of Electrical and Electronics Engineers . IEEE 2019:7–12

    work page 2019

  55. [55]

    Identifying self-admitted technical debt in open source projects using text mining

    Huang Q, Shihab E, Xia X, Lo D, Li S. Identifying self-admitted technical debt in open source projects using text mining. Empirical Software Engineering.2018;23(1):418–451. doi: 10.1007/s10664-017-9522-4

    work page doi:10.1007/s10664-017-9522-4 2018

  56. [56]

    SATD detector: A text-mining-based self-Admitted technical debt detection tool

    Liu Z, Huang Q, Xia X, Shihab E, Lo D, Li S. SATD detector: A text-mining-based self-Admitted technical debt detection tool. In: Institute of Electrical and Electronics Engineers . 2018:9–12

    work page 2018

  57. [57]

    Automatic Classifying Self-Admitted Technical Debt Using N-Gram IDF

    Wattanakriengkrai S, Srisermphoak N, Sintoplertchaikul S, et al. Automatic Classifying Self-Admitted Technical Debt Using N-Gram IDF. In: . 2019-Decem.Institute of Electrical and Electronics Engineers . IEEE 2019:316–322

    work page 2019

  58. [58]

    Identification of Self-Admitted Technical Debt Using Enhanced Feature Selection Based on Word Embedding

    Flisar J, Podgorelec V. Identification of Self-Admitted Technical Debt Using Enhanced Feature Selection Based on Word Embedding. IEEE Access.2019;7:106475–106494. doi: 10.1109/ACCESS.2019.2933318

    work page doi:10.1109/access.2019.2933318 2019

  59. [59]

    Maipradit R, Lin B, Nagy C, et al. Automated Identification of On-hold Self-Admitted Technical Debt.Proceedings - 20th IEEE International Working Conference on Source Code Analysis and Manipulation, SCAM 2020.2020;29(6223):54–64. doi: 10.1109/SCAM51674.2020.00011

    work page doi:10.1109/scam51674.2020.00011 2020

  60. [60]

    Beyond the Code: Mining Self-Admitted Technical Debt in Issue Tracker Systems

    Xavier L, Ferreira F, Brito R, Valente MT. Beyond the Code: Mining Self-Admitted Technical Debt in Issue Tracker Systems. In: Association for Computing Machinery . 2020:137–146

    work page 2020

  61. [61]

    Predicting technical debt from commit contents: reproduction and extension with automated feature selection

    Rantala L, Mäntylä M. Predicting technical debt from commit contents: reproduction and extension with automated feature selection. Software Quality Journal.2020;28(4):1551–1579. doi: 10.1007/s11219-020-09520-3

    work page doi:10.1007/s11219-020-09520-3 2020

  62. [62]

    Prevalence, Contents and Automatic Detection of KL-SATD

    Rantala L, Mantyla M, Lo D. Prevalence, Contents and Automatic Detection of KL-SATD. In:Institute of Electrical and Electronics Engineers. 2020:385–388

    work page 2020

  63. [63]

    Self-Admitted Technical Debt classification using LSTM neural network.Advances in Intelligent Systems and Computing.2020;1134(Itng):679–685

    Santos RM, Junior MCR, Mendonça Neto dMG. Self-Admitted Technical Debt classification using LSTM neural network.Advances in Intelligent Systems and Computing.2020;1134(Itng):679–685. doi: 10.1007/978-3-030-43020-7_93

    work page doi:10.1007/978-3-030-43020-7_93 2020

  64. [64]

    Data balancing improves self-admitted technical debt detection

    Sridharan M, Mantyla M, Rantala L, Claes M. Data balancing improves self-admitted technical debt detection. In:Institute of Electrical and Electronics Engineers . IEEE 2021:358–368

    work page 2021

  65. [65]

    DebtHunter: A machine learning-based approach for detecting self-admitted technical debt

    Sala I, Tommasel A, Arcelli Fontana F. DebtHunter: A machine learning-based approach for detecting self-admitted technical debt. ACM International Conference Proceeding Series.2021:278–283. doi: 10.1145/3463274.3464455

    work page doi:10.1145/3463274.3464455 2021

  66. [66]

    Identifying Self-Admitted Technical Debts With Jitterbug: A Two-Step Approach.IEEE Transactions on Software Engineering.2020;48(5):1676–1691

    Yu Z, Fahid FM, Tu H, Menzies T. Identifying Self-Admitted Technical Debts With Jitterbug: A Two-Step Approach.IEEE Transactions on Software Engineering.2020;48(5):1676–1691. doi: 10.1109/TSE.2020.3031401

    work page doi:10.1109/tse.2020.3031401 2020

  67. [67]

    Multiclass Classification for Self-Admitted Technical Debt Based on XGBoost.IEEE Transactions on Reliability.2022;71(3):1309–1324

    Chen X, Yu D, Fan X, Wang L, Chen J. Multiclass Classification for Self-Admitted Technical Debt Based on XGBoost.IEEE Transactions on Reliability.2022;71(3):1309–1324. doi: 10.1109/TR.2021.3087864

    work page doi:10.1109/tr.2021.3087864 2022

  68. [68]

    SATDBailiff-mining and tracking self-admitted technical debt.Science of Computer Programming.2022;213:102693

    AlOmar EA, Christians B, Busho M, et al. SATDBailiff-mining and tracking self-admitted technical debt.Science of Computer Programming.2022;213:102693. doi: 10.1016/j.scico.2021.102693

    work page doi:10.1016/j.scico.2021.102693 2022

  69. [69]

    A two-stage approach for identifying and interpreting self-admitted technical debt.Applied Intelligence

    Yin M, Wang J, Zhu D, Gao C. A two-stage approach for identifying and interpreting self-admitted technical debt.Applied Intelligence. 2023;53(22):26592–26602

    work page 2023

  70. [70]

    Automated Identification and Prioritization of Self-Admitted Technical Debt Using NLP Word Embeddings

    G SMC, R PK. Automated Identification and Prioritization of Self-Admitted Technical Debt Using NLP Word Embeddings. In: IEEE. 2023:963-971

    work page 2023

  71. [71]

    Investigation on Self-Admitted Technical Debt in Open-Source Blockchain Projects

    Pinna A, Lunesu MI, Orrù S, Tonelli R. Investigation on Self-Admitted Technical Debt in Open-Source Blockchain Projects. Future Internet.2023;15(7):232

    work page 2023

  72. [72]

    Towards Realistic SATD Identification through Machine Learning Models: Ongoing Research and Preliminary Results

    Gama E, Paixao M, Cortés MI, Monteiro L. Towards Realistic SATD Identification through Machine Learning Models: Ongoing Research and Preliminary Results. In: ACM. 2024:675–676

    work page 2024

  73. [73]

    Fixme: A github bot for detecting and monitoring on-hold self-admitted technical debt

    Phaithoon S, Wongnil S, Pussawong P, et al. Fixme: A github bot for detecting and monitoring on-hold self-admitted technical debt. In: IEEE. 2021:1257–1261

    work page 2021

  74. [74]

    Self-admitted technical debt in R: detection and causes.Automated Software Engineering.2022;29(2)

    Sharma R, Shahbazi R, Fard FH, Codabux Z, Vidoni M. Self-admitted technical debt in R: detection and causes.Automated Software Engineering.2022;29(2). doi: 10.1007/s10515-022-00358-6

    work page doi:10.1007/s10515-022-00358-6 2022

  75. [75]

    Automatic Detection and Analysis of Technical Debts in Peer-Review Documentation of R Packages

    Khan JY, Uddin G. Automatic Detection and Analysis of Technical Debts in Peer-Review Documentation of R Packages. In: Institute of Electrical and Electronics Engineers . 2022:765–776

    work page 2022

  76. [76]

    Self-admitted technical debt classification using natural language processing word embeddings

    Sabbah AF, Hanani AA. Self-admitted technical debt classification using natural language processing word embeddings. International Journal of Electrical and Computer Engineering.2023;13(2):2142–2155. doi: 10.11591/ijece.v13i2.pp2142-2155

    work page doi:10.11591/ijece.v13i2.pp2142-2155 2023

  77. [77]

    Automatic identification of self-admitted technical debt from four different sources.Empirical Software Engineering.2023;28(3):65

    Li Y, Soliman M, Avgeriou P. Automatic identification of self-admitted technical debt from four different sources.Empirical Software Engineering.2023;28(3):65

    work page 2023

  78. [78]

    Neural network-based detection of self-Admitted technical debt: From performance to explainability.ACM Transactions on Software Engineering and Methodology.2019;28(3)

    Ren X, Xing Z, Xia X, Lo D, Wang X, Grundy J. Neural network-based detection of self-Admitted technical debt: From performance to explainability.ACM Transactions on Software Engineering and Methodology.2019;28(3). doi: 10.1145/3324916

    work page doi:10.1145/3324916 2019

  79. [79]

    Detecting and Explaining Self-Admitted Technical Debts with Attention-based Neural Networks

    Wang X, Liu J, Li L, Chen X, Liu X, Wu H. Detecting and Explaining Self-Admitted Technical Debts with Attention-based Neural Networks. In:Institute of Electrical and Electronics Engineers . 2020:871–882

    work page 2020

  80. [80]

    Santos RM, Santos IM, Rodrigues MC, Mendonça Neto dMG. Long term-short memory neural networks and word2vec for self-admitted technical debt detection.ICEIS 2020 - Proceedings of the 22nd International Conference on Enterprise Information Systems. 2020;2(May):157–165. doi: 10.5220/0009796001570165

    work page doi:10.5220/0009796001570165 2020

Showing first 80 references.