An Empirical Comparison of Model Validation Techniques for Defect Prediction Models
The DOI matches its Crossref record; 132 of the 132 references it cites were found in a registry or an index.
Abstract
Defect prediction models help software quality assurance teams to allocate their limited resources to the most defect-prone modules. Model validation techniques, such as$k$-fold cross-validation, use historical data to estimate how well a model will perform in the future. However, little is known about how accurate the estimates of model validation techniques tend to be. In this paper, we investigate the bias and variance of model validation techniques in the domain of defect prediction. Analysis of 101 public defect datasets suggests that 77 percent of them are highly susceptible to producing unstable results– - selecting an appropriate model validation technique is a critical experimental design choice. Based on an analysis of 256 studies in the defect prediction literature, we select the 12 most commonly adopted model validation techniques for evaluation. Through a case study of 18 systems, we find that single-repetition holdout validation tends to produce estimates with 46-229 percent more bias and 53-863 percent more variance than the top-ranked model validation techniques. On the other hand, out-of-sample bootstrap validation yields the best balance between the bias and variance of estimates in the context of our study. Therefore, we recommend that future defect prediction studies avoid single-repetition holdout validation, and instead, use out-of-sample bootstrap validation.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Biases may be balanced with variance of performance estimates using out-of-sample bootstrap techniques [139].”
From Operationalising ethics in artificial intelligence for healthcare: a framework for AI developers (2023) · cites this paper as [139]
Citations
Cites
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538
Statistical Power Analysis for the Behavioral Sciences
Muller, K. (1989) · Technometrics · doi:10.1080/00401706.1989.10488618 · Cited by 99,167
Statistical power analysis for the behavioral sciences
(1990) · Computers, Environment and Urban Systems · doi:10.1016/0198-9715(90)90050-4 · Cited by 65,269
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1994) · doi:10.1201/9780429246593 · Cited by 40,500
The Elements of Statistical Learning
Hastie, T., Tibshirani, R., & Friedman, J. (2009) · Springer Series in Statistics · doi:10.1007/978-0-387-84858-7 · Cited by 25,970
The Elements of Statistical Learning
Hastie, T., Friedman, J., & Tibshirani, R. (2001) · Springer Series in Statistics · doi:10.1007/978-0-387-21606-5 · Cited by 24,162
The meaning and use of the area under a receiver operating characteristic (ROC) curve.
Hanley, J. A., & McNeil, B. J. (1982) · Radiology · doi:10.1148/radiology.143.1.7063747 · Cited by 21,856
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1993) · doi:10.1007/978-1-4899-4541-9 · Cited by 21,230
An Analysis of Transformations
Box, G. E. P., & Cox, D. R. (1964) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1964.tb00553.x · Cited by 15,052
A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection
Kohavi, R. (1995) · Zenodo · doi:10.5281/zenodo.19712698 · Cited by 10,909
Research works printed but not linked: 9
- 10.29172/7c2a6982-6d72-4cd8-bba6-2fccb06a7011 · DOI · Paper page
- 10.2307/2289236 · doi:10.2307/2289236 · Paper page
- Jelihovschi, 2014, The ScottKnott Clustering Algorithm
- 10.1067/mod.2000.109032 · DOI · Paper page
- Meyer, 2014, E1071: Misc functions of the department of statistics (E1071), TU Wien
- 10.1109/ICSE.2005.1553571 · DOI · Paper page
- 2013, R Core Team, R: A language and environment for statistical computing, R foundation for statistical computing, Vienna, Austria
- 10.1007/978-3-031-01899-2 · doi:10.1007/978-3-031-01899-2 · Paper page
- Tantithamthavorn, 2015, ScottKnottESD: An R package of the Scott-Knott effect size difference (ESD) test
Cited by
Items 301–400 of 575
When less is more: on the value of “co-training” for semi-supervised software defect predictors
Majumder, S., Chakraborty, J., & Menzies, T. (2024) · Empirical Software Engineering · doi:10.1007/s10664-023-10418-4 · Cited by 20
Huang, Z., Shao, Z., Fan, G., Yu, H., Yang, X., & Yang, K. (2022) · Journal of Computer Science and Technology · doi:10.1007/s11390-021-1596-1 · Cited by 11
“In order to rank the performance with e(cid:11)ect size awareness, we also involve the SK-ESD [42] test.”
MethodInfluentialAntiCopyPaster: Extracting Code Duplicates As Soon As They Are Introduced in the IDE
AlOmar, E. A., Ivanov, A., Kurbatova, Z., Golubev, Y., Mkaouer, M. W., Ouni, A., Bryksin, T., Nguyen, L., Kini, A., & Thakur, A. (2022) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1145/3551349.3559537 · Cited by 11
“In order to evaluate the performance of the algorithms, we use out-of-sample bootstrap validation since this validation technique yields the best balance between the bias and variance in comparison to single-repetition holdout validation [18].”
MethodA novel Bayes defect predictor based on information diffusion function
Wu, Y., Huang, S., Ji, H., Zheng, C., & Bai, C. (2017) · Knowledge-Based Systems · doi:10.1016/j.knosys.2017.12.015 · Cited by 11
Optimizing Prediction Intervals by Tuning Random Forest via Meta-ValidationPreprint
Bayley, S., & Falessi, D. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1801.07194 · Cited by 18
“There are many validation techniques that can be applied and no consensus on which technique is the most effective [4, 13, 15, 47, 48, 65, 86].”
InfluentialOperationalizing validity of empirical software engineering studies
Härtel, J., & Lämmel, R. (2023) · Empirical Software Engineering · doi:10.1007/s10664-023-10370-3 · Cited by 13
Zhao, Y. D., Huang, Z., Gong, L., Zhu, Y., Yu, Q., & Gao, Y. (2023) · IET Software · doi:10.1049/2023/6293074 · Cited by 17
“Tantithamthavorn et al. [65] conducted an empirical study of the various model validation techniques.”
BackgroundCode Multiview Hypergraph Representation Learning for Software Defect Prediction
Qiu, S., Huang, M., Liang, Y., Peng, C., & Yuan, Y. (2024) · IEEE Transactions on Reliability · doi:10.1109/tr.2024.3393415 · Cited by 10
Tunkel, S., & Herbold, S. (2022) · Empirical Software Engineering · doi:10.1007/s10664-022-10224-4 · Cited by 11
“According to Tantithamthavorn et al. (2017), this should lead to an unbiased estimator of the performance means.”
BackgroundStudying backers and hunters in bounty issue addressing process of open source projects
Zhou, J., Wang, S., Zhang, H., Chen, T.-H. P., & Hassan, A. E. (2021) · Empirical Software Engineering · doi:10.1007/s10664-021-09979-z · Cited by 12
Self-citationGonçalves, P. W., Çalıklı, G., Serebrenik, A., & Bacchelli, A. (2023) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3579471 · Cited by 28
“The Scott-Knott ESD has been successfully used in the past to analyze software engineering data [12, 13, 22, 68].”
MethodCombining Data Preprocessing Methods With Imputation Techniques for Software Defect Prediction
Kakkar, M., Jain, S., Bansal, A., & Grover, P. S. (2018) · International Journal of Open Source Software and Processes · doi:10.4018/ijossp.2018010101 · Cited by 11
Sayagh, M., & Hassan, A. E. (2020) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2020.2973997 · Cited by 11
“…We calculate the accuracy at 1, 5, 10, and 20 on 2,000 different bootstrap samples to mitigate the errors within the 2,062 config-related questions and to produce statistically robust conclusions similar to prior work [20], [22], since bootstrap resampling produces more accurate estimates [23].”
MethodSelf-citationHOPE: Software Defect Prediction Model Construction Method via Homomorphic Encryption
Yu, C., Ding, Z., & Chen, X. (2021) · IEEE Access · doi:10.1109/access.2021.3078265 · Cited by 11
“In our empirical study, we use Accuracy [31] and AUC [22] (area under ROC curve) as performance measures to measure the model performance because these performance measures are the most concerning measures in the previous studies on software defect prediction [46], [47].”
MethodAn ensemble‐based predictive mutation testing approach that considers impact of unreached mutants
Aghamohammadi, A., & Mirian‐Hosseinabadi, S. (2021) · Software Testing Verification and Reliability · doi:10.1002/stvr.1784 · Cited by 14
“This is due to the fact that software metrics play an important role in research areas such as defect prediction [56-61], mutation testing [14,27,62,63] and testability [64].”
BackgroundEmpirical Analysis of Data Sampling-Based Ensemble Methods in Software Defect Prediction
Balogun, A. O., Odejide, B. J., Bajeh, A. O., Alanamu, Z. O., Usman-Hamza, F. E., Adeleke, H. O., Mabayoje, M. A., & Yusuff, S. R. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-10548-7_27 · Cited by 10
A pattern-based validation method for the credibility evaluation of simulation models
Laili, Y., Li, Z., & Luo, Y. (2019) · SIMULATION · doi:10.1177/0037549719856100 · Cited by 11
Synergy between Machine/Deep Learning and Software Engineering: How Far Are We?Preprint
Wang, S., Huang, L., Ge, J., Zhang, T., Feng, H., Li, M., Zhang, H., & Ng, V. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.05515 · Cited by 9
“so the opposite, ML/DL for SE. Nevertheless, these two studies perfectly complement our SLR, which can encourage SE researchers to address those challenges. Many SLRs, surveys and comparative studies [159, 160, 161, 9, 162, 52, 53, 163] have focused on investigating the use of ML/DL in software defect prediction. Besi…”
BackgroundA Formal Explainer for Just-In-Time Defect Predictions
Yu, J., Fu, M. C., Ignatiev, A., Tantithamthavorn, C. K., & Stuckey, P. J. (2024) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3664809 · Cited by 19
“15:”
Self-citationAwal, M. A., & Roy, C. K. (2024) · Journal of Systems and Software · doi:10.1016/j.jss.2024.112159 · Cited by 16
“In our study, we follow commonly used hyperparameter settings to train ML models [8, 9, 12, 23, 50, 51, 52, 53].”
MethodExplainable artificial intelligence for machine learning prediction of bandgap energies
Masuda, T., & Tanabe, K. (2024) · Journal of Applied Physics · doi:10.1063/5.0226151 · Cited by 12
Evaluation of Tree-Based Voting Algorithms in Water Quality Classification Prediction
Li-li, L., & Wei, J. (2024) · Sustainability · doi:10.3390/su162310634 · Cited by 11
Context-aware code summary generation
Su, C., Bansal, A., Huang, Y., Li, T. J.-J., & McMillan, C. (2025) · Journal of Systems and Software · doi:10.1016/j.jss.2025.112580 · Cited by 26
An Empirical Study of Self-Admitted Technical Debt in Machine Learning Software
Bhatia, A., Khomh, F., Adams, B., & Hassan, A. E. (2026) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3785001 · Cited by 40
“SATD is both introduced and removed significantly earlier in ML projects in comparison to non-ML projects.”
Self-citationMutant Reduction Evaluation: What is There and What is Missing?
Zhang, P., Wang, Y., Liu, X., Li, Y., Yang, Y., Wang, Z., Zhou, X., Chen, L., & Zhou, Y. (2022) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3522578 · Cited by 9
“0) [39] to compute for each strategy a rank (as shown in brackets, a lower rank is better).”
BackgroundPredicting Crash Fault Residence via Simplified Deep Forest Based on A Reduced Feature Set
Zhao, K., Liu, J., Zhou, X., Li, L., Yan, M. L., Yu, J., & Zhou, Y. (2021) · IEEE International Conference on Program Comprehension (ICPC) · doi:10.1109/icpc52881.2021.00031 · Cited by 9
Yang, X., Yu, H., Fan, G., Huang, Z., Yang, K., & Zhou, Z. (2021) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-92635-9_25 · Cited by 13
Defect prediction model using transfer learning
Malhotra, R., & Meena, S. (2022) · Soft Computing · doi:10.1007/s00500-022-06846-x · Cited by 9
Distance Metric Recommendation for k-Means Clustering: A Meta-Learning Approach
Gonzales, M. E. M., Uy, L. C., Sy, J. A. L., & Cordel, M. O. (2022) · IEEE Region 10 Conference (TENCON) · doi:10.1109/tencon55691.2022.9978037 · Cited by 11
“Applying the Scott-Knott effect size difference test [34] showed that the mean RA of using the meta-learning model is”
MethodA Method-Level Defect Prediction Approach Based on Structural Features of Method-Calling Network
Yang, F., Xu, H., Xiao, P., Zhong, F., & Zeng, G. (2023) · IEEE Access · doi:10.1109/access.2023.3239266 · Cited by 12
“In the software defect prediction contexts [10], within-project software defect prediction [3], [13] uses historical data from the same project and version to build training set and test set.”
MethodAn introductory guide for conducting psychological research with big data.
Vezzoli, M., & Zogmaister, C. (2023) · Psychological Methods · doi:10.1037/met0000513 · Cited by 12
Ahasanuzzaman, M., Oliva, G. A., & Hassan, A. E. (2023) · Empirical Software Engineering · doi:10.1007/s10664-023-10421-9 · Cited by 11
“The SK-ESD is a method of multiple comparison that leverages a hierarchical clustering to partition the set of treatment values (e.g., medians or means) into statistically distinct groups with non-negligible difference [40].”
MethodSelf-citationLiu, W., Yue, Y., Chen, X., Gu, Q., Zhao, P., Liu, X., & Zhao, J. (2024) · Information and Software Technology · doi:10.1016/j.infsof.2024.107510 · Cited by 9
A Hybrid Nonlinear Manifold Detection Approach for Software Defect Prediction
Ghosh, S., Rana, A., & Kansal, V. (2018) · International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito.2018.8748788 · Cited by 9
“Though series of research work have so far been made for software defect prediction at early stage [5], [6], [15], [21] but still many issues and challenges are still left in this field.”
BackgroundZhao, Y., Wang, Y., Zhang, D., & Gong, Y. (2022) · Expert Systems · doi:10.1111/exsy.12977 · Cited by 9
“ScottKnottESD is selected in this paper, which is a mean comparison approach that leverages a hierarchical clustering to partition the set of treatment means into statistically distinct groups with non-negligible difference (Tantithamthavorn et al., 2016, 2017; Y. Zhang, Jin, et al., 2020).”
MethodEvaluating the adaptive selection of classifiers for cross-project bug prediction
Nucci, D. D., Palomba, F., & Lucia, A. D. (2018) · International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering (RAISE) · doi:10.1145/3194104.3194112 · Cited by 9
“A key decision in this context was the selection of an appropriate validation strategy [41].”
BackgroundResultInfluentialOn the effectiveness of developer features in code smell prioritization: A replication study
Huang, Z., Yu, H., Fan, G., Shao, Z., Zhou, Z., & Li, M. (2024) · Journal of Systems and Software · doi:10.1016/j.jss.2024.111968 · Cited by 15
Enhancing Performance Bug Prediction Using Performance Code Metrics
Zhao, G., Georgiou, S., Hassan, S., Zou, Y., Truong, D., & Corbin, T. (2024) · International Conference on Mining Software Repositories (MSR) · doi:10.1145/3643991.3644920 · Cited by 17
“We apply the out-of-sample bootstrap technique [3, 32, 81, 82] to create training and testing datasets for each project.”
MethodHow do Software Engineering Researchers Use GitHub? An Empirical Study of Artifacts & Impact
Alrashedy, K., & Binjahlan, A. (2024) · IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM) · doi:10.1109/scam63643.2024.00021 · Cited by 13
Fu, L., Wang, T., Ouyang, M., Zhao, L. L., & Yin, X. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.111473 · Cited by 8
An Empirical Study on Multi-Source Cross-Project Defect Prediction Models
Liu, X., Li, Z., Zou, J., & Tong, H. (2022) · Asia-Pacific Software Engineering Conference (APSEC) · doi:10.1109/apsec57359.2022.00044 · Cited by 8
“In this paper, we adopt Scott-Knott ESD test [39] to investigate whether the performance difference among these MSCPDP models is statistically significant.”
MethodKim, E., Baik, J., & Ryu, D. (2021) · IEEE International Conference on Software Quality, Reliability and Security (QRS) · doi:10.1109/qrs54544.2021.00061 · Cited by 8
Krasniqi, R., & Do, H. (2023) · Empirical Software Engineering · doi:10.1007/s10664-022-10280-w · Cited by 10
“We select these techniques based on recommendations and an extensive empirical study performed by a recent work of Tantithamthavorn et al. (2016).”
MethodAn unsupervised feature selection approach for actionable warning identification
Ge, X., Fang, C., Liu, J., Qing, M., Li, X., & Zhao, Z. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120152 · Cited by 8
Comparing Hyperparameter Optimization in Cross- and Within-Project Defect Prediction: A Case Study
Öztürk, M. M. (2018) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-018-3564-9 · Cited by 9
Dynamic stacking ensemble for cross-language code smell detection
Aljamaan, H. (2024) · PeerJ Computer Science · doi:10.7717/peerj-cs.2254 · Cited by 7
“With this 10 repetition of the 10 fold cross validation, we ensure that we produce unbiased model estimates with low variance ( Tantithamthavorn et al., 2016 ).”
MethodMulti-view learning based on product and process metrics for software defect prediction
Sun, Y., Wu, D., Wu, D., Jing, X., & Sun, Y. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06288-6 · Cited by 7
Multi-Class Skin Disease Detection Using Deep Learning Hybrid Method
Faris, M., Qayyum, R., Pepe, C., Ali, M. F., & Zanoli, S. M. (2025) · International Carpathian Control Conference · doi:10.1109/iccc65605.2025.11022791 · Cited by 10
“This division ensured a balanced approach, supporting effective model training, validation, and evaluation [19].”
Studying the Practices of Logging Exception Stack Traces in Open-Source Software Projects
Li, H., Zhang, H., Wang, S., & Hassan, A. E. (2021) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2021.3129688 · Cited by 15
“Therefore, we use another SK clustering to further group the metrics into ranked groups by statistically comparing their ranks in different projects ( a.k.a. , double Scott-Knott clustering [19], [57]).”
MethodSelf-citationJoint Instance and Feature Adaptation for Heterogeneous Defect Prediction
Ren, Y., Liu, B., & Wang, S. (2023) · IEEE Transactions on Reliability · doi:10.1109/tr.2023.3305356 · Cited by 9
A Hybrid Multi-Filter Wrapper Feature Selection Method for Software Defect Predictors
Balogun, A. O., Shuib, B., Abdulkadir, S. J., & Hashim, A. S. (2019) · Cited by 7
Prioritizing code documentation effort: Can we do it simpler but better?
Liu, S., Guo, Z., Li, Y., Lu, H., Chen, L., Xu, L., Zhou, Y., & Xu, B. (2021) · Information and Software Technology · doi:10.1016/j.infsof.2021.106686 · Cited by 10
Deployment of a change‐level software defect prediction solution into an industrial setting
Eken, B., Tufan, S., Tunaboylu, A., Guler, T., Atar, R., & Tosun, A. (2021) · Journal of Software Evolution and Process · doi:10.1002/smr.2381 · Cited by 10
“For comparing different models' performance in Sections 6.2 and 6.3, we applied the Scott-Knott Effect Size Difference (ESD) ¶ test implemented by Tantithamthavorn et al. 68 This test provides a statistical comparison among the performance achieved by different variations of the online prediction models.”
MethodInfluentialSoftware change‐proneness prediction based on deep learning
Zhu, X., Li, N., & Wang, Y. (2022) · Journal of Software Evolution and Process · doi:10.1002/smr.2434 · Cited by 11
“To verify whether the differences between the different methods discussed above are significant, we use the ScottKnottESD (Scott – Knott effect size difference) 67 method to test.”
MethodAn empirical study of issue-link algorithms: which issue-link algorithms should we use?
Kondo, M., Kashiwa, Y., Kamei, Y., & Mizuno, O. (2022) · Empirical Software Engineering · doi:10.1007/s10664-022-10120-x · Cited by 7
“, the Scott-Knott ESD test [80]) and the effect size (e.”
BackgroundMethodInfluentialBessghaier, N., Sayagh, M., Ouni, A., & Mkaouer, M. W. (2023) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3607186 · Cited by 7
“To make our evaluation statistically robust, we leverage 100 bootstrap samples for training and testing our models similarly to prior studies [29, 38, 73– 75, 77, 78].”
MethodNi, C., Xu, X., Yang, K., & Lo, D. (2023) · International Conference on Mining Software Repositories (MSR) · doi:10.1109/msr59073.2023.00072 · Cited by 7
“We apply the Scott-Knott ESD (SKESD) test on the feature importance scores following existing works [4, 21, 22, 50, 53].”
MethodEmpirical validation of feature selection techniques for cross-project defect prediction
Malhotra, R., & Meena, S. (2023) · International Journal of Systems Assurance Engineering and Management · doi:10.1007/s13198-023-02051-7 · Cited by 8
A Review of Audio-Visual Speech Recognition
Seong, T. W., & Ibrahim, M. Z. (2018) · Journal of Telecommunication Electronic and Computer Engineering (JTEC) · Cited by 7
Failure of One, Fall of Many: An Exploratory Study of Software Features for Defect Prediction
dos Santos, G. E., & Figueiredo, E. (2020) · IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM) · doi:10.1109/scam51674.2020.00016 · Cited by 10
“In order to statistically test the soundness of the base-line results, we applied a Scott-Knott Effect Size Difference (ESD) test [36], [37].”
MethodSemi-supervised Heterogeneous Defect Prediction with Open-source Projects on GitHub
Sun, Y., Jing, X., Wu, F., Dong, X., Sun, Y., & Wang, R. (2021) · International Journal of Software Engineering and Knowledge Engineering · doi:10.1142/s0218194021500273 · Cited by 8
Quantifying the advantage of multimodal data fusion for survival prediction in cancer patientsPreprint
Νικολάου, Ν., Salazar, D., RaviPrakash, H., Gonçalves, M., Mulla, R., Burlutskiy, N., Markuzon, N., & Jacob, E. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.01.08.574756 · Cited by 7
Balanced Adversarial Tight Matching for Cross‐Project Defect Prediction
Jiang, S., Zhang, J., Guo, F., Ouyang, T., & Li, J. (2024) · IET Software · doi:10.1049/2024/1561351 · Cited by 8
“Additionally, this paper applies a variant of the Scott-Knott effect size difference (ESD) test [41] to quantify the effect size of the discrepancy between the two methods.”
MethodAn Empirical Evaluation of Ensemble Models for Python Code Smell Detection
Rao, R. S., Dewangan, S., & Mishra, A. (2025) · Applied Sciences · doi:10.3390/app15137472 · Cited by 6
Mobile Application Online Cross-Project Just-in-Time Software Defect Prediction Framework
Jiang, S., He, Z., Chen, Y., Zhang, M., & Ma, L. (2024) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3664607 · Cited by 7
“Additionally, in order to analyze the significance of our method and all comparative methods, we adopted a statistical test called Scott-Knott effect size difference (Scott-Knott ESD) [34].”
Rabello, J. P., Cavalcante, P. E. D. S., Leme, J., Dias, V. A. T., Barrence, F. A. C., Guardalini, L. G. O., Bernardino, T. C., Nunes, R., Barros, I. H. A. S., Tonso, A., et al. (2024) · Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy · doi:10.1016/j.saa.2024.125217 · Cited by 6
Bagri, R., Rajpal, A., & Kumar, N. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130797 · Cited by 6
Assi, M., Hassan, S., Georgiou, S., & Zou, Y. (2023) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3593802 · Cited by 8
“The SK-ESD clusters the topics of issue reports in a way where the intra-group difference in severity level is negligible and the inter-group difference is non-negligible.”
MethodInfluentialData Science with Semantic Technologies: Application to Information Systems Development
Sassi, S. B., & Yanes, N. (2023) · Journal of Computer Information Systems · doi:10.1080/08874417.2023.2220294 · Cited by 7
“110 and software quality (Rossi et al. 111 ; Tantithamthavorn et al.))”
MethodLearning Feature Representations from Change Dependency Graphs for Defect Prediction
Loyola, P., & Matsuo, Y. (2017) · Proceedings/Proceedings - International Symposium on Software Reliability Engineering · doi:10.1109/issre.2017.30 · Cited by 6
“The problem of the representation of the data has been present in the literature, with recent focus on the aggregation effects [26] and model validation [27].”
BackgroundAn empirical study of software change classification with imbalance data‐handling methods
Zhu, X., Niu, B., Whitehead, E. J., & Sun, Z. (2018) · Software Practice and Experience · doi:10.1002/spe.2606 · Cited by 7
Predicting Defects with Latent and Semantic Features from Commit Logs in an Industrial Setting
Eken, B., Atar, R., Sertalp, S., & Tosun, A. (2019) · IEEE/ACM International Conference on Automated Software Engineering Workshops (ASEW) · doi:10.1109/asew.2019.00038 · Cited by 6
“Scott-Knott ESD test clusters the treatments into statistically different groups by comparing their means [37].”
MethodApplication of DPSIR model in prediction of ecological sustainable development capacity in Bohai Sea
Li, X., Bao, J., Sun, J., & Wang, J. (2021) · Arabian Journal of Geosciences · doi:10.1007/s12517-021-06866-1 · Cited by 7
Manchala, P., Tiwari, A., & Bisi, M. (2024) · Soft Computing · doi:10.1007/s00500-024-10379-w · Cited by 5
“2–5: We employed the Wilcoxon signed rank (WSR) (Turhan et al. 2009; Tantithamthavorn et al. 2016) test to evaluate the statistical performance differences between the two approaches.”
Testing Machine Learning and Deep Learning Systems: Achievements and Challenges
Albelali, S., & Ahmed, M. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10276-w · Cited by 8
Zou, Y., Wang, H., Lv, H., Zhao, S., & Tian, H. (2024) · International Journal of Software Engineering and Knowledge Engineering · doi:10.1142/s0218194024500165 · Cited by 5
Khan, A., Ali, A., Khan, J., Ullah, F., & Faheem, M. (2025) · IEEE Access · doi:10.1109/access.2025.3558913 · Cited by 7
“Bootstrap validation supports estimating FS rankings variability by producing several resampled datasets [11], [38].”
InfluentialLiu, B., Zhang, H., Yang, L., Dong, L., Shen, H., & Song, K. (2020) · International Conference on Evaluation and Assessment in Software Engineering (EASE) · doi:10.1145/3383219.3383222 · Cited by 6
“We did not deliberately make the two validation processes have the same training or testing set as Tantithamthavorn et al. did [52] since whether or not the test set will introduce future data is the major difference between the two methods.”
MethodTeReKG: A temporal collaborative knowledge graph framework for software team recommendation
Ruenin, P., Choetkiertikul, M., Supratak, A., & Tuarob, S. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111492 · Cited by 11
Akhal, H. E., Yahya, A. B., & Alaoui, A. E. B. E. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109646 · Cited by 5
Automated Recommendation of Extracting Local Variable Refactorings
Jiang, Y., Chi, X., Zhang, Y., Ji, W., Li, G., Wang, W., Xia, Y., Zhang, L., & Liu, H. (2025) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3715110 · Cited by 5
“In the preceding steps, the proposed approach validates that the given expression can be extracted (as a local variable) and it has constructed an initial refactoring solution (i.e., which expressions should be extracted and where the new variable should be declared).”
Cross Domain Few-Shot Line-Level Defect Prediction in Open Software Development via Meta Learning
Li, Z., Guo, S., Zhang, L., Li, H., Yang, Z., Qiao, N., & Ma, Q. (2025) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3572334 · Cited by 6
“The Scott-Knott ESD test [78] is an enhanced version of the standard Scott-Knott test [79], which accounts for effect size.”
Pitfalls Analyzer: Quality Control for Model-Driven Data Science Pipelines
Rajbahadur, G. K., Oliva, G. A., Hassan, A. E., & Juergen Dingel, A. (2019) · ACM/IEEE International Conference on Model Driven Engineering Languages and Systems (MODELS) · doi:10.1109/models.2019.00-19 · Cited by 7
Self-citationAn Empirical Study on Progressive Sampling for Just-in-Time Software Defect Prediction.
Yang, X., Yu, H., Fan, G., Yang, K., & Shi, K. (2019) · Cited by 5
Amasaki, S., Aman, H., & Yokogawa, T. (2022) · Empirical Software Engineering · doi:10.1007/s10664-021-10103-4 · Cited by 5
“The CPDP approach configurations were ranked using Scott-Knott ESD test (Tantithamthavorn et al. 2017).”
MethodUsability of Computer-Aided Translation Software Based on Deep Learning Algorithms
Liu, J. (2022) · Advances in Multimedia · doi:10.1155/2022/9047053 · Cited by 5
“+e top layer represents the unit feature vector of the hidden layer h, the bottom layer is the unit vector of the visible layer v, and the link weight w is the model parameter [22, 23].”
Heterogeneous Defect Prediction through Joint Metric Selection and Matching
Chen, H., Jing, X., & Xu, B. (2021) · IEEE International Conference on Software Quality, Reliability and Security (QRS) · doi:10.1109/qrs54544.2021.00048 · Cited by 4
WIFLF: An approach independent of the target project for cross‐project defect prediction
Cui, C., Liu, B., & Wang, S. (2022) · Journal of Software Evolution and Process · doi:10.1002/smr.2497 · Cited by 5
“Thus, to avoid results bias, we employ Wilcoxon Signed-Rank test 25 with 95% significance level, Cohen's d effect size test 26 and SK-ESD test 27 to assess the performance between our approach and the benchmarks.”
MethodA Deep Learning Based Model for Defect Prediction in Intra-Project Software
Sekaran, K., & Annabel, L. S. P. (2023) · International Conference on Trends in Electronics and Informatics (ICOEI) · doi:10.1109/icoei56765.2023.10126014 · Cited by 6
Techniques Based on Data Science for Software Processes: A Systematic Literature Review
Carpio, Á. F. D., & Angarita, L. B. (2018) · Communications in computer and information science · doi:10.1007/978-3-030-00623-5_2 · Cited by 5
Li, Z., Zhu, W., Zhang, H., Miao, Y., & Ren, J. (2025) · Automated Software Engineering · doi:10.1007/s10515-025-00510-y · Cited by 4
“To mitigate this concern, we employed the out-of-sample bootstrap validation technique (Tantithamthavorn et al. 2017), which is known for providing performance estimates with minimal bias and high stability.”
Enhancing Python Code Smell Detection with Heterogeneous Ensembles
Sandouka, R., & Aljamaan, H. (2025) · International Journal of Software Engineering and Knowledge Engineering · doi:10.1142/s0218194025500287 · Cited by 4
Aragón-Jurado, J. M., Bangash, A. A., Dorronsoro, B., Ali, K., Hindle, A., & Ruiz, P. (2025) · Sustainable Computing Informatics and Systems · doi:10.1016/j.suscom.2025.101166 · Cited by 4
A Preliminary Evaluation of CPDP Approaches on Just-in-Time Software Defect Prediction
Amasaki, S., Aman, H., & Yokogawa, T. (2021) · Euromicro Conference on Software Engineering and Advanced Applications (SEAA) · doi:10.1109/seaa53835.2021.00042 · Cited by 4
Recent Advances in Audio-Visual Speech Recognition: Deep Learning Perspective
Pawar, D. R., & Yannawar, P. L. (2023) · Advances in intelligent systems research/Advances in Intelligent Systems Research · doi:10.2991/978-94-6463-196-8_31 · Cited by 4
Towards graph-anonymization of software analytics data: empirical study on JIT defect prediction
Malik, A., Adams, B., & Hassan, A. E. (2024) · Empirical Software Engineering · doi:10.1007/s10664-024-10464-6 · Cited by 6
Self-citationIntegrating Soil Sensing and Rainfall Modelling for Precision Crop Recommendation Systems
Reddy, T. V., Saikumar, K., Reddy, R. A., Chary, V. B., Srinija, P., & Reddy, T. H. V. (2024) · IEEE International Conference for Convergence in Technology (I2CT) · doi:10.1109/i2ct61223.2024.10543793 · Cited by 4
“Comprehending these relationships played a pivotal role in the process of feature selection and model construction, guaranteeing the incorporation of relevant variables to enable precise soil state classification by the machine learning models [3].”
Zhang, N., Zhu, K., & Zhu, D. (2024) · IET Software · doi:10.1049/2024/5358773 · Cited by 4
Sample analysis and multi-label classification for malicious sample datasets
Xie, J., Li, S., Yun, X., Si, C., & Yin, T. (2025) · Computer Networks · doi:10.1016/j.comnet.2024.110999 · Cited by 4
Interpretable Machine Learning Models for Predicting Cesarean Delivery in Class III Obese Cohorts
Bennett, R., Pierce, S., & Razzaghi, T. (2025) · IEEE Access · doi:10.1109/access.2025.3547992 · Cited by 4
“We also employ the Non-Parametric Scott-Knott (NPSK) Effect Size Difference test [61] in order to compare and rank the results of each model.”
Metadata
- Authors
- Tantithamthavorn, C., McIntosh, S., Hassan, A. E., & Matsumoto, K.
- Affiliations
- Nara Institute of Science and Technology (JP)
- Queen's University (CA)
- Venue
- IEEE Transactions on Software Engineering
- Published
- 2017-01-01
- Volume
- 43
- Issue
- 1
- Pages
- 1-18
- ISSN
- 0098-5589 (ISSN-L)
- 1939-3520
- OpenAlex
- W2474835145
- Cited by
- 574
- References
- 129
- Citations by year
- 2026: 39 · 2025: 53 · 2024: 69 · 2023: 56 · 2022: 58 · 2021: 93 · 2020: 63 · 2019: 75 · 2018: 45 · 2017: 18
- FWCI
- 106.29
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Software Engineering (publishedVersion)
- Funders
- JSPS
- Advancing Strategic International Networks to Accelerate the Circulation of Talented Researchers
- Interdisciplinary Global Networks for Accelerating Theory and Practice in Software Ecosystem
- JSPS Fellows (16J03360)
- Natural Sciences and Engineering Research Council of Canada (NSERC)
- Topics
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Software › Software Reliability and Analysis Research
- Physical Sciences › Computer Science › Information Systems › Software Engineering Techniques and Practices
- Keywords
- fault prediction, model validation, k-fold cross-validation, software quality assurance, defect dataset
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.