Evaluating the Quality of Machine Learning Explanations: A Survey on Methods and Metrics
Open accessThe DOI matches its Crossref record; 88 of the 88 references it cites were found in a registry or an index.
Abstract
The most successful Machine Learning (ML) systems remain complex black boxes to end-users, and even experts are often unable to understand the rationale behind their decisions. The lack of transparency of such systems can have severe consequences or poor uses of limited valuable resources in medical diagnosis, financial decision-making, and in other high-stake domains. Therefore, the issue of ML explanation has experienced a surge in interest from the research community to application domains. While numerous explanation methods have been explored, there is a need for evaluations to quantify the quality of explanation methods to determine whether and to what extent the offered explainability achieves the defined objective, and compare available explanation methods and suggest the best explanation from the comparison for a specific task. This survey paper presents a comprehensive overview of methods proposed in the current literature for the evaluation of ML explanations. We identify properties of explainability from the review of definitions of explainability. The identified properties of explainability are used as objectives that evaluation metrics should achieve. The survey found that the quantitative metrics for both model-based and example-based explanations are primarily used to evaluate the parsimony/simplicity of interpretability, while the quantitative metrics for attribution-based explanations are primarily used to evaluate the soundness of fidelity of explainability. The survey also demonstrated that subjective measures, such as trust and confidence, have been embraced as the focal point for the human-centered evaluation of explainable systems. The paper concludes that the evaluation of ML explanations is a multidisciplinary research topic. It is also not possible to define an implementation of evaluation metrics, which can be applied to all explanation methods.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The cited source says what the claim attributes to it.
In the citing paper
“They found that some properties (clarity, broadness, and completeness) are still in shortage of appropriate metrics, and so is the class of explanations that are example-based.”
From Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review (2021) · cites this paper as [141]
In the citing paper
“It is not possible to come up with quantitative metrics that are applicable for all interpretability methods [169].”
From Transparency of deep neural networks for medical image analysis: A review of interpretability methods (2022) · cites this paper as [169]
In the citing paper
“They conclude their survey by stating that”
From Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review (2021) · cites this paper as [141]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707 · cited 2×
“One way to objectively evaluate explanations is to verify whether the explanation mechanism satisfies (or does not satisfy) certain axioms, or properties [34].”
Background"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204 · cited 1×
“[39] explained an ensemble of models, such as neural networks with a single model effectively.”
BackgroundCausality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653
Rudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 1×
“Explanation is also a domain-specific note, and there cannot be an all-purpose explanation [22], and different types of explanations might be useful.”
BackgroundPeeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
Adadi, A., & Berrada, M. (2018) · IEEE Access · doi:10.1109/access.2018.2870052 · Cited by 6,294 · cited 2×
“The machine learning explainability has been reviewed thoroughly in recent years [10,14,17,23,28,31,32].”
BackgroundLipton, Z. C. (2018) · Queue · doi:10.1145/3236386.3241340 · Cited by 5,427
Explanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302
Axiomatic attribution for deep networks
Sundararajan, M., Taly, A., & Yan, Q. (2017) · arXiv · doi:10.48550/arxiv.1703.01365 · Cited by 4,087 · cited 4×
“…Main effect complexity [46] Level of (dis)agreement [82] Attribution-based explanations Monotonicity [80] Non-sensitivity [80,83], Sensitivity [84,85] Effective complexity [80] Remove and retrain [86] Recall of important features [33] Implementation invariance [85] Selectivity [87] Continuity…”
BackgroundInfluentialTowards A Rigorous Science of Interpretable Machine Learning
Doshi-Velez, F., & Kim, B. (2017) · arXiv · doi:10.48550/arxiv.1702.08608 · Cited by 3,121 · cited 4×
“The advantages of application-grounded and human-grounded evaluations are that they can provide direct and strong evidence of success of explanations [9].”
BackgroundMethodInfluential
Research works printed but not linked: 5
- Zhou, J., and Chen, F. (2018). Human and Machine Learning: Visible, Explainable, Trustworthy and Transparent, Springer.
- Lage, I., and Doshi-Velez, F. (2020, January 17). Human-in-the-Loop Learning of Interpretable and Intuitive Representations. Proceedings of the ICML Workshop on Human Interpretability in Machine Learning, Vienna, Austria.
- Becker, B., Kohavi, R., and Sommerfield, D. (2001). Visualizing the Simple Bayesian Classifier. Information Visualization in Data Mining and Knowledge Discovery, Morgan Kaufmann Publishers Inc.
- Poursabzi-Sangdeh, F., Goldstein, D.G., Hofman, J.M., Vaughan, J.W., and Wallach, H. (2019). Manipulating and Measuring Model Interpretability. arXiv.
- Yang, M., and Kim, B. (2019). Benchmarking Attribution Methods with Relative Feature Importance. arXiv, 1907.
Cited by
Items 201–300 of 676
Gallo, M., Gallo, M., Nenutil, R., Němeček, M., Holub, P., & Brázdil, T. (2023) · New Biotechnology · doi:10.1016/j.nbt.2023.09.008 · Cited by 20
Pandey, C., Angryk, R. A., Georgoulis, M. K., & Aydin, B. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-45275-8_38 · Cited by 16
“To address this, various methods [34] have been proposed to interpret CNNs.”
MethodA Trustworthy View on Explainable Artificial Intelligence Method Evaluation
Ding, L., Liu, Y., Huang, J., & Wang, Z. (2023) · Computer · doi:10.1109/mc.2022.3233806 · Cited by 15
“At both levels, existing work 16 highlights quantified metrics that have been defined to measure XAI explanation results toward the progress of developing trustworthy AI.”
BackgroundMachine Learning Based Diagnosis of Lumpy Skin Disease
Gambhir, S., Khanna, S., & Malhotra, P. (2023) · doi:10.1109/icaia57370.2023.10169125 · Cited by 15
Schmitt, V., Csomor, B. P., Meyer, J., Villa-Arenas, L. F., Jakob, C., Polzehl, T., & Möller, S. (2024) · ACM International Workshop on Multimedia AI against Disinformation (MAD) · doi:10.1145/3643491.3660283 · Cited by 19
“Therefore, our research expands on the theoretical taxonomy, integrating relevant constructs drawn from prior interdisciplinary studies in the HCI and XAI domains [28, 35, 60, 65], thereby enabling an empirical exploration of these dimensions [50].”
MethodAnalyzing and Forecasting E‐Commerce Adoption Drivers Among SMEs: A Machine Learning Approach
Daoud, Y., & Kammoun, A. (2024) · Human Behavior and Emerging Technologies · doi:10.1155/2024/7747136 · Cited by 22
Velmurugan, M., Ouyang, C., Xu, Y., Sindhgatta, R., Wickramanayake, B., & Moreira, C. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109772 · Cited by 18
Extraction of human understandable insight from machine learning model for diabetes prediction
Assegie, T. A., Karpagam, T., Mothukuri, R., Tulasi, R. L., & Engidaye, M. F. (2022) · Bulletin of Electrical Engineering and Informatics · doi:10.11591/eei.v11i2.3391 · Cited by 17
Effects of Fairness and Explanation on Trust in Ethical AI
Angerschmid, A., Theuermann, K., Holzinger, A., Chen, F., & Zhou, J. J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-14463-9_4 · Cited by 19
Self-citationDuval, J., Vidal, L. T., Segura, E. M., Li, Y., & Wærn, A. (2023) · International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) · doi:10.1145/3597638.3608421 · Cited by 16
“In these contexts, predictive power and accuracy are critically important [76].”
BackgroundUnderstanding Disparities in Post Hoc Machine Learning Explanation
Mhasawade, V., Rahman, S., Haskell-Craig, Z., & Chunara, R. (2024) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3630106.3659043 · Cited by 19
Generating Robust Counterfactual Witnesses for Graph Neural Networks
Qiu, D., Wang, M., Khan, A., & Wu, Y. (2024) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde60146.2024.00259 · Cited by 14
“…should indicate “invariant” representative structures for similar graphs that fall into the same group, i.e., be “robust” to small changes of the graphs, and be both “factual” (that preserves the result of classification) and “counterfactual” (which flips the result if removed from G ) [5].”
A review of machine learning methods for non-invasive blood pressure estimation
Pal, R., Le, J., Rudas, Á., Chiang, J. N., Williams, T. M., Alexander, B., Joosten, A., & Cannesson, M. P. (2024) · Journal of Clinical Monitoring and Computing · doi:10.1007/s10877-024-01221-7 · Cited by 16
“Despite their utilization, a stan - dardized framework for clearly reporting results is still lacking [ 42 ].”
Explainable AI for Audio and Visual Affective Computing: A Scoping Review
Johnson, D. S., Hakobyan, O., Paletschek, J., & Drimalla, H. (2024) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2024.3505269 · Cited by 27
“Explanation evaluation is a challenging task as there is often no ground truth for quality assessment [57].”
InfluentialLiu, J., Wu, X., Liu, S., & Gong, S. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.126063 · Cited by 19
The effectiveness of using artificial intelligence in clinical medicine
Иванович, К. Д., & Lamotkin, A. I. (2025) · FARMAKOEKONOMIKA Modern Pharmacoeconomics and Pharmacoepidemiology · doi:10.17749/2070-4909/farmakoekonomika.2025.287 · Cited by 14
Shi, H., Xu, T., Li, X., Gao, Q., Xiong, Z., Xia, J., & Yue, Z. (2025) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2025.103101 · Cited by 16
“While increasing research focuses on interpretability evaluation [48, 49], an evaluation framework for explaining bipartite network interactions remained underdeveloped.”
Dong, Y., Sun, Y., Wang, D., & Waygood, E. O. D. (2025) · Transportation Research Part F Traffic Psychology and Behaviour · doi:10.1016/j.trf.2025.04.019 · Cited by 16
An ensemble-driven machine learning framework for enhanced water quality classification
Singh, P. D., Hasija, T., Bharany, S., Naeem, H., Rao, B. C., Hussen, S., & Rehman, A. U. (2025) · Discover Sustainability · doi:10.1007/s43621-025-01467-4 · Cited by 14
“Performance matrices to measure the efficiency of the model are accuracy, precision, recall, F1-score, Jaccard index and Matthews Correlation Coefficient (MCC) [66, 67].”
Huang, X., Xu, F., Zhu, W., Yao, L., He, J., Su, J., Zhao, W., & Hu, H. (2025) · BMC Medical Imaging · doi:10.1186/s12880-025-01813-y · Cited by 17
“Accuracy, sensitivity, specificity, and odds ratio were derived from the confusion matrix [34, 35] to evaluate classification quality.”
Papantonis, I., & Belle, V. (2025) · Frontiers in Computer Science · doi:10.3389/fcomp.2025.1560448 · Cited by 14
Shapley values for cluster importance: How clusters of the training data affect a prediction
Brandsæter, A., & Glad, I. K. (2022) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-022-00896-3 · Cited by 14
“This is the case for any explanations or interpretation technique, and no single set of evaluation metrics can be applied to all explanation methods (Zhou et al 2021).”
Zhuparris, A., de Goede, A. A., Yocarini, I. E., Kraaij, W., Groeneveld, G. J., & Doll, R. J. (2023) · Sensors · doi:10.3390/s23115243 · Cited by 16
Natural Language Processing and Text Mining (Turning Unstructured Data into Structured)
Bagheri, A., Giachanou, A., Mosteiro, P. J., & Verberne, S. (2023) · Springer eBooks · doi:10.1007/978-3-031-36678-9_5 · Cited by 21
Zhu, A. Y. F. (2024) · Technology in Society · doi:10.1016/j.techsoc.2024.102599 · Cited by 13
XRL-Bench: A Benchmark for Evaluating and Comparing Explainable Reinforcement Learning Techniques
Xiong, Y., Hu, Z., Huang, Y., Wu, R., Guan, K., Fang, X., Jiang, J. T., Zhou, T., Hu, Y., Liu, H., et al. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671595 · Cited by 16
“Fidelity pertains to the explanation’s correlation with the actual rationale for agent decision-making [1, 29, 59], while stability ensures that the explanations remain largely consistent despite minor perturbations to the input [2, 14, 42].”
Hybrid AI and semiconductor approaches for power quality improvement
Chinthaginjala, R., Srinivasulu, A., Agrawal, A., Kim, T., Tera, S. P., & Ahmad, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-11116-5 · Cited by 16
How Explainable Is Explainability? Towards Better Metrics for Explainable AI
Pawlicka, A., Pawlicki, M., Kozik, R., Kurek, W., & Choraś, M. (2023) · Springer proceedings in complexity · doi:10.1007/978-3-031-44721-1_52 · Cited by 14
GastroVRG: Enhancing early screening in gastrointestinal health via advanced transfer features
Islam, M. S., Rony, M. A. T., & Sultan, T. (2024) · Intelligent Systems with Applications · doi:10.1016/j.iswa.2024.200399 · Cited by 13
Impact of example-based XAI for neural networks on trust, understanding, and performance
Perlmutter, M., Gifford, R. C., & Krening, S. (2024) · International Journal of Human-Computer Studies · doi:10.1016/j.ijhcs.2024.103277 · Cited by 13
Hassan, S., Li, Q., Zubair, M., Alsowail, R. A., & Qureshi, M. A. (2024) · Sustainability · doi:10.3390/su16145901 · Cited by 13
Usability Evaluation of Lecturer Information System in ITB STIKOM Bali
Dwipayana, I. G. M. S., Candiasa, . M., & Dewi, L. J. E. (2025) · SinkrOn · doi:10.33395/sinkron.v9i1.14315 · Cited by 12
Villarreal-Torres, H., Ángeles-Morales, J., Rodriguez, W. J. M., Andrade-Girón, D., Carreño-Cisneros, E., Cano-Mejía, J., Mejía-Murillo, C., Boscán, M., Reyes, G. F., & Cruz-Cruz, O. (2023) · ICST Transactions on Scalable Information Systems · doi:10.4108/eetsis.3489 · Cited by 13
“…box) to understand its internal functioning by a person; This is where interpretability plays a vitally important role due to the lack of transparency and thus avoiding negative consequences, producing correct answers for the wrong reasons in high-risk areas (Rudin et al., 2021; Zhou et al., 2021).”
BackgroundEvaluating Model Explanations without Ground Truth
Rawal, K., Fu, Z., Delaney, E. D., & Russell, C. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732219 · Cited by 15
“Real world situations lack access to an oracle to provide ground-truth explanations e ∗ [41, 58].”
Predictability and Comprehensibility in Post-Hoc XAI Methods: A User-Centered Analysis
Jalali, A., Haslhofer, B., Kriglstein, S., & Rauber, A. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-37717-4_46 · Cited by 14
Ribeiro, J. M., Andrade, P., Carvalho, M., Silva, C., Ribeiro, B., & Roque, L. (2022) · Mathematics · doi:10.3390/math10091604 · Cited by 11
“[12] presents a comprehensive overview of methods for the evaluation of ML explanations and Linardatos et al.”
MethodArtificial intelligence in pharmacovigilance: Do we need explainability?
Hauben, M. (2022) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.5501 · Cited by 11
“Other recent methods assess individual feature importance by iteratively tracing back from outputs to inputs via reverse propagation of activations and weights in a neural network (layer-wise relevance propagation).(1,4,10) Testing with concept activations vectors provide explanations based on higher level human-intui…”
BackgroundMethodInfluentialNew optimization models for optimal classification trees
Alès, Z., Huré, V., & Lambert, A. (2023) · Computers & Operations Research · doi:10.1016/j.cor.2023.106515 · Cited by 18
“There is currently no clear metric to measure interpretability but rather a consensus on which models are (or are not) interpretables [Zhou et al., 2021].”
Hachicha, M., Halima, R. B., & Frikha, T. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-10877-w · Cited by 10
“To evaluate each ML model, we present the calculation of various evaluation metrics [33] mainly the mean squared error (MSE), mean absolute error (MAE) and coefficient of determination (R2)[34].”
Dugăeșescu, A., & Florea, A. M. (2025) · Neural Computing and Applications · doi:10.1007/s00521-025-11282-7 · Cited by 10
“Explain-able artificial intelligence (XAI) is a research area that has gained substantial momentum in recent years [1–3], developing approaches and methodologies to elucidate the intricate workings of machine learning models, particularly deep neural networks.”
Global Trends in Artificial Intelligence. Challenges, Opportunities, and Prospects
Zgurovsky, M. Z. (2025) · Cybernetics and Systems Analysis · doi:10.1007/s10559-025-00790-y · Cited by 12
“Table 6 presents the importance ratings of each challenge on a scale from 1 to 10, where 1 indicates minimal impact and 10 represents a critical level of limitation for modern models [34].”
Hybrid Machine Learning Meta-Model for the Condition Assessment of Urban Underground Pipes
Mohammadagha, M., Najafi, M., Kaushal, V., & Jibreen, A. (2025) · Infrastructures · doi:10.3390/infrastructures10110282 · Cited by 15
“Model evaluation is conducted via multifaceted metrics [31], including accuracy [32], precision [33], recall [34], and F1 score [35], supplemented by visualizations and feature importance analyses [36].”
Radočaj, D., Jurišić, M., Plašćak, I., & Galić, L. (2025) · Agronomy · doi:10.3390/agronomy15112495 · Cited by 13
Efficient GNN Explanation via Learning Removal-based Attribution
Rong, Y., Wang, G., Feng, Q., Liu, N., Liu, Z., Kasneci, E., & Hu, X. (2024) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3685678 · Cited by 13
Incorporating Machine Learning into Sociological Model-Building
Verhagen, M. D. (2024) · Sociological Methodology · doi:10.1177/00811750231217734 · Cited by 17
Alarcon, G. M., & Willis, S. (2023) · Hawaii International Conference on System Sciences (HICSS) · doi:10.24251/hicss.2023.134 · Cited by 18
An Experimental Investigation into the Evaluation of Explainability Methods for Computer Vision
Stassin, S., Englebert, A., Nanfack, G., Albert, J., Versbraegen, N., Peiffer, G., Doh, M., Riche, N., Frénay, B., & Vleeschouwer, C. D. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-74630-7_7 · Cited by 12
Concept Distillation in Graph Neural Networks
Magister, L. C., Barbiero, P., Kazhdan, D., Siciliano, F., Ciravegna, G., Silvestri, F., Jamnik, M., & Lió, P. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-44070-0_12 · Cited by 15
XAI Method Properties: A (Meta-)study.Preprint
Schwalbe, G., & Finzel, B. (2021) · arXiv (Cornell University) · Cited by 10
A Framework for Evaluating Post Hoc Feature-Additive ExplainersPreprint
Carmichael, Z., & Scheirer, W. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.08376 · Cited by 12
“Emphasizing some of the arguments made in [42, 60], our framework can alleviate some of the financial and other resource burdens by filtering out explainers, e.”
BackgroundFirdaus, M. P., & Trisnawarman, D. (2025) · MALCOM Indonesian Journal of Machine Learning and Computer Science · doi:10.57152/malcom.v5i1.1744 · Cited by 9
Improving Cloud Efficiency: A Machine Learning-Based Stacking Model for CPU Utilization Prediction
Javeed, A., Borg, A., Grahn, H., Lundberg, L., Patel, D., & Shirinbab, S. (2025) · International Conference on Data Science and Machine Learning Applications (CDMA) · doi:10.1109/cdma61895.2025.00026 · Cited by 11
“Researchers used different validation schemes to assess the performance of the machine learning models, such as cross-validation, holdout validation, shuffle split, [18], [19], etc.”
Otuka, C. I., Cai, D., Ukwuoma, C. C., Luo, S., Yang, Z., Bamisile, O., Ukwuoma, C. D., Otuka, C. U., Otuka, N. O., & Huang, Q. (2025) · Solar Energy · doi:10.1016/j.solener.2025.114175 · Cited by 10
Basic Issues and Challenges on Explainable Artificial Intelligence (XAI) in Healthcare Systems
Oladipo, I. D., Awotunde, J. B., AbdulRaheem, M., & Salihu, S. A. (2022) · Advances in medical technologies and clinical practice book series · doi:10.4018/978-1-6684-3791-9.ch011 · Cited by 10
Classification Algorithms for Liver Epidemic Identification
Makkena, K. R., & Natarajan, K. (2023) · EAI Endorsed Transactions on Pervasive Health and Technology · doi:10.4108/eetpht.9.4379 · Cited by 10
The Blame Problem in Evaluating Local Explanations and How to Tackle It
Rahnama, A. H. A. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-50396-2_4 · Cited by 9
Kebir, S. T., Berrhail, F., & Didi, F. (2024) · Brazilian Journal of Technology · doi:10.38152/bjtv7n2-002 · Cited by 9
Cheng, X., Li, R., Xie, P., Wang, X., Yu, L., Wu, R., Xue, H., & Bi, Y. (2024) · Postharvest Biology and Technology · doi:10.1016/j.postharvbio.2024.113115 · Cited by 9
Predicting nighttime black ice using atmospheric data for efficient winter road maintenance patrols
Jang, J. (2024) · Transportation Research Interdisciplinary Perspectives · doi:10.1016/j.trip.2024.101299 · Cited by 9
Deep learning inspired game-based cognitive assessment for early dementia detection
Maji, P. K., Acharya, S., Paul, P., Chakraborty, S., & Basu, S. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109901 · Cited by 13
Fuzzy Decision‐Making Framework for Evaluating Hybrid Detection Models of Trauma Patients
Hamid, R. A., Zahid, I. A., Albahri, A. S., Albahri, A. S., Alamoodi, A. H., Alzubaidi, L., Sharaf, I. M., Joudar, S. S., Gu, Y., & Al-Qaysi, Z. T. (2025) · Expert Systems · doi:10.1111/exsy.70005 · Cited by 9
“This methodology has to be addressed because various metrics probably possess different levels of significance based on the given application (Zhou et al. 2021).”
Bhaskaran, S. V. (2025) · Computers · doi:10.3390/computers14030106 · Cited by 13
“These values have been obtained from the confusion metrics analysis [37,38].”
No tillage outperforms conventional tillage under arid conditions and following fertilization
Veresoglou, S. D., Chen, J., Du, X., Fu, Q., Geng, Q., Huang, C., Huang, X., Hu, N., Hun, Y., Li, G. C., et al. (2022) · Soil Ecology Letters · doi:10.1007/s42832-022-0145-3 · Cited by 10
“As a rule of thump linear models tend to be more parsimonious that machine learning approaches and can reveal underlying mechanisms (e.g., Zhou et al., 2021), meaning that we can be confident about the sign of the generated coefficients.”
BackgroundEvalAttAI: A Holistic Approach to Evaluating Attribution Maps in Robust and Non-Robust Models
Nielsen, I. E., Ramachandran, R. P., Bouaynaya, N. C., Fathallah‐Shaykh, H. M., & Rasool, G. (2023) · IEEE Access · doi:10.1109/access.2023.3300242 · Cited by 16
“Many different techniques have been proposed in the literature to measure the faithfulness or fidelity [9], [11], [12], [13], [28], [29], [30], [31], [32], [33], [34].”
BackgroundInterpretability and Explainability of Machine Learning Models: Achievements and Challenges
Henriques, J. M. O., Rocha, T., Carvalho, P., Silva, C., & Paredes, S. (2024) · World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany · doi:10.1007/978-3-031-59216-4_9 · Cited by 13
Evaluation of the Shapley Additive Explanation Technique for Ensemble Learning Methods
Assegie, T. A. (2022) · Proceedings of Engineering and Technology Innovation · doi:10.46604/peti.2022.9025 · Cited by 9
“Designing and implementing interpretable ensemble learning methods for heart disease diagnosis has become one of the recent research topics in machine learning [4].”
BackgroundTulloh, R., Syambas, N. R., & Mulyana, E. (2024) · IEEE Student Conference on Research and Development (SCOReD) · doi:10.1109/scored64708.2024.10872655 · Cited by 8
“Machine learning (ML) is a method that facilitates predictive analysis on large, complex datasets through modeling and learned algorithms [10].”
Koçyiğit, E., Rossi, A., Sergeeva, A., Negri-Ribalta, C., Farjami, A., & Lenzini, G. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732129 · Cited by 9
Qiao, X., Cheng, J., Liu, H., Xiao, X., Liu, Q., Zhou, J., Yan, Z., Cheng, C., & Yu, M. (2025) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2025.113212 · Cited by 8
Self-citationAlarcon, G. M., & Capiola, A. (2025) · Frontiers in Computer Science · doi:10.3389/fcomp.2025.1662185 · Cited by 13
“Instead, personality variables such as complacency potential may be active in the cognitive processing because of the low inherent risk (Zhou et al., 2020).”
InfluentialLiu, J., Wu, X., Xie, Y., Tang, Z., Xie, Y., & Gong, S. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121138 · Cited by 11
Phishing Detection Using Machine Learning: A Model Development and Integration
Onih, V. (2024) · International Journal of Scientific and Management Research · doi:10.37502/ijsmr.2024.7403 · Cited by 8
“The evaluation carried out in this study is built on these key metrics: Moreover, the Confusion Matrix is an essential tool for evaluating the efficacy of the model development [26].”
BackgroundMethodInfluentialTowards Improved XAI-Based Epidemiological Research into the Next Potential Pandemic
Khalili, H., & Wimmer, M. A. (2024) · Life · doi:10.3390/life14070783 · Cited by 8
“Elaborating more on different non-deterministic approaches, including ensemble models (to best synthesize the predictions of multiple basic models) [93] and Bayesian neural networks (to infer distributions over the models’ weights and outputs) [155], can enhance future epidemiological pipelines.”
BackgroundLess Is More: The Influence of Pruning on the Explainability of CNNs
Merkle, F., Weber, D., Schöttle, P., Schlögl, S., & Nocker, M. (2025) · IEEE Access · doi:10.1109/access.2025.3569575 · Cited by 11
“Functionally-grounded evaluation metrics, on the other hand, consist of various quantitative metrics to objectively assess the quality of an explanation, or more specifically whether certain explainability axioms are met [95].”
InfluentialTowards a Taxonomy for Explainable AI in Computational Pathology
Müller, H., Kargl, M., Plass, M., Kipperer, B., Brčić, L., Regitnig, P., Geißler, C., Küster, T., Zerbe, N., & Holzinger, A. (2021) · Springer eBooks · doi:10.1007/978-3-030-72188-6_15 · Cited by 9
Self-citationInterpretable Machine Learning in Social Sciences: Use Cases and Limitations
Suvorova, A. V. (2022) · Communications in computer and information science · doi:10.1007/978-3-030-93715-7_23 · Cited by 8
Democratizing Quality-Based Machine Learning Development through Extended Feature Models
d’Aloisio, G., Marco, A. D., & Stilo, G. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-30826-0_5 · Cited by 11
Real-Time Detection of Unrecognized Objects in Logistics Warehouses Using Semantic Segmentation
Carata, S.-V., Ghenescu, M. T., & Mihaescu, R. E. (2023) · Mathematics · doi:10.3390/math11112445 · Cited by 8
Experiential Explanations for Reinforcement Learning
Alabdulkarim, A., Singh, M., Mansi, G., Hall, K., Ehsan, U., & Riedl, M. O. (2025) · Neural Computing and Applications · doi:10.1007/s00521-024-10951-3 · Cited by 13
Quantifying Explainability in NLP and Analyzing Algorithms for Performance-Explainability TradeoffPreprint
Naylor, M., French, C., Terker, S., & Kamath, U. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.05693 · Cited by 13
“There are a handful of metrics which seek to quantify the quality of model explanations (Zhou et al., 2021), however there is no single metric which will adequately cover all aspects of interpretability and apply to all contexts.”
Intention-aware policy graphs: answering what, how, and why in opaque agentsPreprint
Giménez-Ábalos, V., Álvarez-Napagao, S., Tormos, A., Cortés, U., & Vázquez-Salceda, J. (2024) · arXiv (Cornell University) · doi:10.5281/zenodo.13862642 · Cited by 9
A unified approach to extract interpretable rules from tree ensembles via Integer Programming
Bonasera, L., & Carrizosa, E. (2025) · Computers & Operations Research · doi:10.1016/j.cor.2025.107283 · Cited by 9
A multi-AI approach to predicting municipal solid waste generation and recycling demand in Hong Kong
Xu, P. J., & Zheng, H. R. (2025) · Resources Conservation and Recycling · doi:10.1016/j.resconrec.2025.108590 · Cited by 7
PEARL: A Rubric-Driven Multi-Metric Framework for LLM Evaluation
Anghel, C., Anghel, A. A., Pecheanu, E., Crăciun, M., Cocu, A., & Niculita, C. (2025) · Information · doi:10.3390/info16110926 · Cited by 12
Tahir, H., Jabbarpour, M. R., Vo, Q. B., Kowalczyk, R., Barr, J., & Bessell, T. (2025) · IEEE Access · doi:10.1109/access.2025.3639551 · Cited by 7
“…that implement safe online learning algorithms [52], [53], [60] could address this limitation, enabling spacecraft to refine their detection models based on newly observed telemetry patterns [166], [167], [168] while preventing catastrophic forgetting of critical failure modes [24], [71], [72].”
Arslan, R. U., Aksoy, B., & Yapıcı, İ. Ş. (2026) · Scientific Reports · doi:10.1038/s41598-026-35173-6 · Cited by 7
Qureshi, M. A., Noor, A. A., Manzoor, A., Mazhar Qureshi, M. D., Younus, A., & Rashwan, W. (2026) · PLOS One · doi:10.1371/journal.pone.0351473 · Cited by 7
“Although the metrics are not new, we formalize and implement their mathematical definitions for transparency and reproducibility [4,5,8].”
InfluentialCheng, M., Liu, K., Liu, Z. F., Xu, J., Zhang, Z., & Sun, C. (2024) · Remote Sensing · doi:10.3390/rs16173280 · Cited by 7
Mehak, S., Ramos, I. F., Sagar, K., Ramasubramanian, A. K., Kelleher, J. D., Guilfoyle, M., Gianini, G., Damiani, E., & Leva, M. C. (2024) · Frontiers in Robotics and AI · doi:10.3389/frobt.2024.1434351 · Cited by 8
The disagreement dilemma in explainable AI: can bias reduction bridge the gap
Bhardwaj, N., & Parashar, G. (2025) · International Journal of Systems Assurance Engineering and Management · doi:10.1007/s13198-025-02712-9 · Cited by 8
Norouzi, A., Ahmadi, I., Nazari, M., Pourrostami, H., & Hoseini-Kordkheili, H. (2025) · Results in Engineering · doi:10.1016/j.rineng.2025.104478 · Cited by 7
Lalinia, M., Nemati, N. H., Mofazali, P., Gross, J. D., & Samadi, A. (2025) · Materials Today Chemistry · doi:10.1016/j.mtchem.2025.102640 · Cited by 13
Evaluating the feasibility of interpretable machine learning for globular cluster detection
Dold, D., & Fahrion, K. (2022) · Astronomy and Astrophysics · doi:10.1051/0004-6361/202243354 · Cited by 8
Finzel, B., Rieger, I., Kuhn, S., & Schmid, U. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-40837-3_3 · Cited by 7
AI-Powered Network Intrusion Detection: A New Frontier in Cybersecurity
Rachini, A. S., Fares, C., Assaf, M. A., Jamal, B., & Khatoun, R. (2023) · International Arab Conference on Information Technology (ACIT) · doi:10.1109/acit58888.2023.10453733 · Cited by 7
“Performance metrics in ML, such as precision, recall, and F1-Score, are pivotal for evaluating the accuracy and effectiveness of predictive models [13].”
Abid, R., Rizwan, M., Alabdulatif, A., Alnajim, A. M., Alamro, M. A., & Azrour, M. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2024.046880 · Cited by 10
Explainability, Quantified: Benchmarking XAI Techniques
Perotti, A., Borile, C., Miola, A., Nerini, F. P., Baracco, P., & Panisson, A. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-63787-2_22 · Cited by 11
El-Khalek, A. A. A., Balaha, H. M., Mahmoud, A. H., Alghamdi, N. S., Ghazal, M. A., Khalil, A. T., Abo-Elsoud, M. E. A., & El-Baz, A. S. (2024) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi56570.2024.10635727 · Cited by 7
“The study employed four main scaling methods: Stan-dardization, Min-Max Scaling (Normalization), and Max Absolute Scaling [23, 24].”
MethodWalke, F., Bennek, L., & Winkler, T. J. (2025) · Lecture notes in information systems and organisation · doi:10.1007/978-3-031-80122-8_8 · Cited by 7
Ketata, F., Masry, Z. A., Yacoub, S., & Zerhouni, N. (2024) · IEEE Access · doi:10.1109/access.2024.3431691 · Cited by 7
“The authors of [28] discussed explainability in AI as a blend of interpretability (how understandable the explanations are to humans) and fidelity (how accurately these explanations reflect the model’s behavior).”
Background
Metadata
- Authors
- Zhou, J., Gandomi, A. H., Chen, F., & Holzinger, A.
- Affiliations
- University of Technology Sydney (AU)
- Medical University of Graz (AT)
- Corresponding author
- Zhou, Jianlong
- Gandomi, Amir H.
- Venue
- Electronics
- Publisher
- MDPI AG
- Published
- 2021-03-04
- Volume
- 10
- Issue
- 5
- Pages
- 593
- ISSN
- 2079-9292 (ISSN-L)
- OpenAlex
- W3133543405
- Cited by
- 606
- References
- 80
- Citations by year
- 2026: 75 · 2025: 138 · 2024: 179 · 2023: 109 · 2022: 84 · 2021: 20
- FWCI
- 34.65
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Electronics (publishedVersion, cc-by)
- OPUS - Open Publications of UTS Scholars (University of Technology Sydney) (submittedVersion, other-oa)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- University of Technology Sydney
- Austrian Science Fund (P 32554)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- explainable machine learning, explanation quality evaluation, evaluation metrics, explainability, model-based explanation, example-based explanations, interpretability, fidelity, human-centered evaluation, black-box models
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.