Explainable AI: A Review of Machine Learning Interpretability Methods
Open accessThe DOI matches its Crossref record; 170 of the 170 references it cites were found in a registry or an index.
Abstract
Recent advances in artificial intelligence (AI) have led to its widespread industrial adoption, with machine learning systems demonstrating superhuman performance in a significant number of tasks. However, this surge in performance, has often been achieved through increased model complexity, turning such systems into “black box” approaches and causing uncertainty regarding the way they operate and, ultimately, the way that they come to decisions. This ambiguity has made it problematic for machine learning systems to be adopted in sensitive yet critical domains, where their value could be immense, such as healthcare. As a result, scientific interest in the field of Explainable Artificial Intelligence (XAI), a field that is concerned with the development of new methods that explain and interpret machine learning models, has been tremendously reignited over recent years. This study focuses on machine learning interpretability methods; more specifically, a literature review and taxonomy of these methods are presented, as well as links to their programming implementations, in the hope that this survey would serve as a reference point for both theorists and practitioners.
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
Figures








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 source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“While valuable, these techniques can sometimes be inconsistent or fail to capture the true decision-making process of highly complex models, potentially leading to a false sense of interpretability66.”
From Bias recognition and mitigation strategies in artificial intelligence healthcare applications (2025) · cites this paper as [66]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“These methods facilitate a deeper understanding of the model’s internal logic, support debugging and optimization processes, and help identify key variables that most strongly influence students’ behavioral intentions toward GenAI adoption (Linardatos et al., 2020).”
From A two-step structural equation modeling and explainable machine learning framework for understanding university students’ adoption of generative AI: balancing intrinsic motivations and perceived risks (2026) · cites this paper as Linardatos et al., 2020
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Yet, there is no common taxonomy neither within one field (such as data science) nor between different fields (law and data science).”
From Transparency of AI in Healthcare as a Multilayered System of Accountabilities: Between Legal Requirements and Technical Limitations (2022) · cites this paper as Linardatos et al., 2021
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Despite the substantial amount of research devoted to solving the black-box issue of AI, many scholars in the field agree that a clear relevant taxonomy is missing (Doshi-Velez and Kim, 2017; Adadi and Berrada, 2018; Linardatos et al., 2021).”
From Transparency of AI in Healthcare as a Multilayered System of Accountabilities: Between Legal Requirements and Technical Limitations (2022) · cites this paper as Linardatos et al., 2021
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Prior research on XAI across industries, has also found that the perceived quality of explanations is closely tied to their actionability and practical relevance for end users [8].”
From Explainable AI in hospital clinical decision support systems: A scoping review of healthcare professionals’ perspectives (2026) · cites this paper as [8]
Experimental — accuracy not yet measured
5 more claims about this paper
Create a free account to see every claim other papers make about this one.
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
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“An improvement over the initial method was proposed in [24], where the well-known Krizhevsky network [25] was utilised in order to outperform state-of-the-art saliency models by a large margin, increasing the amount of explained information by 67% when compared to state-of-the art.”
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 1×
“On the one hand, there are the so called black-box models, which include deep learning [2] and ensembles [8–10].”
Greedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731 · cited 3×
“[59] PDPbox InterpretML Skater PH G Agnostic tab 589.”
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
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 3×
“The breakDown method, as proposed in [62], is similar to SHAP [48], as both, based on the conditioned responses of a black-box model, attempt to attribute them proportionally to the input features.”
Classification and Regression by randomForest
Liaw, A. I., & Wiener, M. C. (2007) · Cited by 18,594
"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
Visualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 8×
“[44] showed that, while the Deconvolution [32], Guided BackPropagation [31], and LRP [37] methods help in interpreting deep neural networks, they do not produce the theoretically correct interpretation, even in the simplest neural network setting; a linear model developed while using data that were produced by a linea…”
InfluentialLearning Deep Features for Discriminative Localization
Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., & Torralba, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.319 · Cited by 11,428 · cited 2×
“In many cases, this is achieved through exploiting the gradient information flowing through the layers of the network, Grad-CAM [35], a direct extension of [34], being a prime and most influential example in terms of citations per year.”
Research works printed but not linked: 3
- Luss, R., Chen, P.Y., Dhurandhar, A., Sattigeri, P., Zhang, Y., Shanmugam, K., and Tu, C.C. (2019). Generating contrastive explanations with monotonic attribute functions. arXiv.
- Van Looveren, A., and Klaise, J. (2019). Interpretable counterfactual explanations guided by prototypes. arXiv.
- Wang, X., Jin, H., and He, K. (2019). Natural language adversarial attacks and defenses in word level. arXiv.
Cited by
Items 301–400 of 1,016
Shobayo, O., & Saatchi, R. (2025) · Diagnostics · doi:10.3390/diagnostics15091072 · Cited by 53
How Validation Methodology Influences Human Activity Recognition Mobile Systems
Bragança, H. L. D. S., Colonna, J. G., Oliveira, H. A. B. F., & Souto, E. (2022) · Sensors · doi:10.3390/s22062360 · Cited by 65
Artificial Intelligence-Enabled Mode-Locked Fiber Laser: A Review
Ma, Q., & Yu, H. (2023) · Nanomanufacturing and Metrology · doi:10.1007/s41871-023-00216-3 · Cited by 56
“Explainable AI [105] can provide new ways to evaluate the uncertainty of AI models, enabling the discovery of new laws of physics and the expansion of human knowledge.”
BackgroundBordekar, H., Cersullo, N., Brysch, M., Philipp, J., & Hühne, C. (2023) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-023-02272-4 · Cited by 58
“Secondly, these models are complex in nature, making them challenging to explain or interpret (Linardatos et al., 2020).”
Legal implications of automated suspicious transaction monitoring: enhancing integrity of AI
Türkşen, U., Benson, V., & Adamyk, B. (2024) · Journal of Banking Regulation · doi:10.1057/s41261-024-00233-2 · Cited by 81
“ML differs from previous expert systems using an inductive approach, allowing computers to discover rules through correlation and classification using large datasets [12].”
Javed, M. F., Khan, M., Fawad, M., Alabduljabbar, H., Najeh, T., & Gamil, Y. (2024) · Scientific Reports · doi:10.1038/s41598-024-65255-2 · Cited by 53
Soladoye, A. A., Aderinto, N., Osho, D., & Olawade, D. B. (2025) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2025.106093 · Cited by 65
Solanke, A. A. (2022) · Forensic Science International: Digital Investigation · doi:10.1016/j.fsidi.2022.301403 · Cited by 52
Condition-based maintenance using machine learning and role of interpretability: a review
Sharma, J., Mittal, M. L., & Soni, G. (2022) · International Journal of Systems Assurance Engineering and Management · doi:10.1007/s13198-022-01843-7 · Cited by 75
Zhao, F., Miao, F., Wu, Y., xiong, Y., Gong, S., & Sun, D. (2023) · Gondwana Research · doi:10.1016/j.gr.2023.09.019 · Cited by 52
Applications of natural language processing at emergency department triage: A narrative review
Stewart, J. E., Lü, J., Goudie, A., Arendts, G., Meka, S., Freeman, S., Walker, K., Sprivulis, P., Sanfilippo, F. M., Bennamoun, M., et al. (2023) · PLoS ONE · doi:10.1371/journal.pone.0279953 · Cited by 70
Panda, C., Mishra, A. K., Dash, A. K., & Nawab, H. (2022) · International Journal of Crashworthiness · doi:10.1080/13588265.2022.2074643 · Cited by 58
“Alternative model agnostic interpretability techniques such as local interpretable model-agnostic explanations (LIME), deterministic LIME (DLIME), partial dependence plot (PDP) and anchors [34] can be experimented in future studies.”
BackgroundInfluentialFuzzy Cognitive Maps: Their Role in Explainable Artificial Intelligence
Apostolopoulos, I. D., & Groumpos, P. P. (2023) · Applied Sciences · doi:10.3390/app13063412 · Cited by 62
Hanna, M. G., Olson, N. H., Zarella, M. D., Dash, R. C., Herrmann, M. D., Furtado, L. V., Stram, M. N., Raciti, P. M., Hassell, L. A., Mays, A., et al. (2023) · Archives of Pathology & Laboratory Medicine · doi:10.5858/arpa.2023-0042-cp · Cited by 58
Miao, J., Zhang, X., Zhang, G., Wei, T., Zhao, Y., Ma, W., Chen, Y., Li, Y., & Wang, Y. (2024) · CATENA · doi:10.1016/j.catena.2024.107848 · Cited by 56
BMB-LIME: LIME with modeling local nonlinearity and uncertainty in explainability
Hung, Y.-H., & Lee, C. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111732 · Cited by 61
Agent-based modeling in cancer biomedicine: applications and tools for calibration and validation
Cogno, N., Axenie, C., Bauer, R., & Vavourakis, V. (2024) · Cancer Biology & Therapy · doi:10.1080/15384047.2024.2344600 · Cited by 62
“Indeed, ML approaches that enable explainability (or inter-pretability) exist, as described in the work of Linardatos et al. 98 In view of the trend of in silico models’ rising complexity and the number of parameters they consist of, then comes the need to rigorously assess ‘ robustness ’ and ‘ sensitivity ’ of the mo…”
Visualization obesity risk prediction system based on machine learning
Du, J., Yang, S., Zeng, Y., Ye, C., Chang, X., & Wu, S. (2024) · Scientific Reports · doi:10.1038/s41598-024-73826-6 · Cited by 56
“This posed challenges to formulating intervention measures to reduce adverse events and greatly limited the application of research results 17,18 .”
BackgroundExplainable AI for DeepFake Detection
Mansoor, N., & Iliev, A. I. (2025) · Applied Sciences · doi:10.3390/app15020725 · Cited by 75
Balancing innovation and integrity: the role of AI in research and scientific writing
BaHammam, A. (2023) · Nature and Science of Sleep · doi:10.2147/nss.s455765 · Cited by 50
“10 Another concern, especially in academic circles, is the reliability of AI-generated text.”
BackgroundValente, F., Paredes, S., Henriques, J. M. O., Rocha, T., Carvalho, P., & Morais, J. (2022) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-022-00821-8 · Cited by 62
“The first is based on the explanation of the importance of each individual feature (inputs) with respect to the prediction outcome; the second relies on the creation of an interpretable model (surrogate model) founded on the ML model previously obtained (Burkart and Huber 2021; Linardatos et al. 2021).”
BackgroundMethodLiang, D., & Xue, F. (2023) · Computers in Industry · doi:10.1016/j.compind.2023.103883 · Cited by 58
“The missing or uninterpreted step-by-step details in these ‘black box’ ML models lead to untrustworthiness for industrial practitioners and regulators who are sensitive to risks in healthcare, legislation (e.g., European Union’s GDPR), and critical engineering (Linardatos et al. 2020).”
Interpretable machine learning for predicting chronic kidney disease progression risk
Zheng, J., Li, X., Zhu, J., Guan, S., Zhang, S.-X., & Wang, W.-M. (2024) · Digital Health · doi:10.1177/20552076231224225 · Cited by 62
“19 Interpretable ML represents a nascent but active fi eld with immense potential to bridge the gap between predictive analytics and clinical implementation.”
BackgroundArtificial intelligence in groundwater management: Innovations, challenges, and future prospects
Shaikh, M., & Birajdar, F. S. (2024) · International Journal of Science and Research Archive · doi:10.30574/ijsra.2024.11.1.0105 · Cited by 58
Optimization Schedule Schemes for Charging Electric Vehicles: Overview, Challenges, and Solutions
Al-Alwash, H., Borcoci, E., Vochin, M., Balapuwaduge, I. A. M., & Li, F. Y. (2024) · IEEE Access · doi:10.1109/access.2024.3371890 · Cited by 54
“The study in [71] presented an extensive and in-depth identification analysis, and provided an comparison of various ML interpretability methods, including Pre-model, In-model, Post-model; Intrinsic, Post-hoc; Model-specific , and Model-agnostic .”
ResultXAI-AV: Explainable Artificial Intelligence for Trust Management in Autonomous Vehicles
Mankodiya, H., Obaidat, M. S., Gupta, R., & Tanwar, S. (2021) · International Conference on Communications, Computing, Cybersecurity, and Informatics (CCCI) · doi:10.1109/ccci52664.2021.9583190 · Cited by 53
Benchmarking Molecular Feature Attribution Methods with Activity Cliffs
Jiménez-Luna, J., Skalic, M., & Weskamp, N. (2022) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.1c01163 · Cited by 47
Novel Vision Transformer–Based Bi-LSTM Model for LU/LC Prediction—Javadi Hills, India
Mohanrajan, S. N., & Loganathan, A. (2022) · Applied Sciences · doi:10.3390/app12136387 · Cited by 53
Kharchenko, V. S., Fesenko, H. V., & Illiashenko, O. A. (2022) · Sensors · doi:10.3390/s22134865 · Cited by 58
Ten quick tips for biomarker discovery and validation analyses using machine learning
Díaz‐Uriarte, R., de Lope, E. G., Giugno, R., Fröhlich, H., Nazarov, P. V., Nepomuceno-Chamorro, I. A., Rauschenberger, A., & Glaab, E. (2022) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1010357 · Cited by 67
“A systematic review of those and further methods can be found in [147,148].”
Kawakura, S., Hirafuji, M., Ninomiya, S., & Shibasaki, R. (2022) · European Journal of Agriculture and Food Sciences · doi:10.24018/ejfood.2022.4.6.348 · Cited by 58
A Machine Learning Explainability Tutorial for Atmospheric Sciences
Flora, M. L., Potvin, C. K., McGovern, A., & Handler, S. L. (2023) · Artificial Intelligence for the Earth Systems · doi:10.1175/aies-d-23-0018.1 · Cited by 77
Unraveling the Black Box: A Review of Explainable Deep Learning Healthcare Techniques
Murad, N. Y., Hasan, M. H., Azam, M. H., Yousuf, N., & Yalli, J. S. (2024) · IEEE Access · doi:10.1109/access.2024.3398203 · Cited by 61
“This technique entails transforming the complicated knowledge embedded inside the DNNs into a more interpretable representation, allowing users seeking insights into the model’s decision-making process to gain a better understanding and transparency [60].”
MethodExplainable AI Frameworks: Navigating the Present Challenges and Unveiling Innovative Applications
Sharma, N. A., Chand, R. R., Buksh, Z., Ali, A. B. M. S., Hanif, A., & Beheshti, A. (2024) · Algorithms · doi:10.3390/a17060227 · Cited by 60
“This method enhances the transparency of CNN-based models, enabling a better understanding of the image classification processes [18,28,49] 4.1.7.”
BackgroundMethodInfluentialA Guide to Feature Importance Methods for Scientific Inference
Ewald, F. K., Bothmann, L., Wright, M. N., Bischl, B., Casalicchio, G., & König, G. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-63797-1_22 · Cited by 77
“However, PIMP is a general procedure and has broader applicability across various FI methods [36,43].”
BackgroundUrban Flood Depth Prediction and Visualization Based on the XGBoost-SHAP Model
Liu, Y., Wang, H., Guan, X., Meng, Y. H., & Xu, H. (2024) · Water Resources Management · doi:10.1007/s11269-024-04020-6 · Cited by 53
Geetha, G. K., & Sim, S. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104572 · Cited by 66
Karami, H., Derakhshani, A., GhasemiGol, M., Fereidouni, M., Miri‐Moghaddam, E., Baradaran, B., Tabrizi, N. J., Najafi, S., Solimando, A. G., Marsh, L. M., et al. (2021) · Journal of Clinical Medicine · doi:10.3390/jcm10163567 · Cited by 53
Laios, A., Kalampokis, E., Johnson, R. L., Thangavelu, A., Tarabanis, C., Nugent, D., & Jong, D. D. (2022) · Journal of Personalized Medicine · doi:10.3390/jpm12040607 · Cited by 54
“To explain the predictive models, we proposed the Shap-ley Additive explanations (SHAP) values as a unified measure of feature importance [28].”
Real-World and Regulatory Perspectives of Artificial Intelligence in Cardiovascular Imaging
Wellnhofer, E. (2022) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2022.890809 · Cited by 56
“Thus a recent review titled: “Explainable AI: A Review of Machine Learning Interpretability Methods” (46).”
BackgroundA quantitative approach for the comparison of additive local explanation methods
Doumard, E., Aligon, J., Escriva, E., Excoffier, J.-B., Monsarrat, P., & Soulé-Dupuy, C. (2022) · Information Systems · doi:10.1016/j.is.2022.102162 · Cited by 59
Machine learning for small molecule drug discovery in academia and industry
Volkamer, A., Riniker, S., Nittinger, E., Lanini, J., Grisoni, F., Evertsson, E., Rodríguez-Pérez, R., & Schneider, N. (2023) · Artificial Intelligence in the Life Sciences · doi:10.1016/j.ailsci.2022.100056 · Cited by 66
Possibilities and ethical issues of entrusting nursing tasks to robots and artificial intelligence
Ibuki, T., Ibuki, A., & Nakazawa, E. (2023) · Nursing Ethics · doi:10.1177/09697330221149094 · Cited by 60
Liu, M., Wen, Z., & Su, H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122022 · Cited by 51
Annuzzi, G., Apicella, A., Arpaïa, P., Bozzetto, L., Criscuolo, S., Benedetto, E. D., Pesola, M., & Prevete, R. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3348334 · Cited by 53
“…white-box models, but also by the critical importance of proactively addressing concerns related to discrimination and biases within datasets and model training [33], in order to prevent disparities and promote equitable representation of the patients’ population in AI-powered systems.”
BackgroundMethodDeep Learning Approaches for Stress Detection: A Survey
Kyrou, M., Kompatsiaris, I., & Petrantonakis, P. C. (2024) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2024.3455371 · Cited by 66
OCT-based diagnosis of glaucoma and glaucoma stages using explainable machine learning
Hasan, M. M., Phu, J., Wang, H., Sowmya, A., Kalloniatis, M., & Meijering, E. H. W. (2025) · Scientific Reports · doi:10.1038/s41598-025-87219-w · Cited by 59
“Shapley values are useful to calculate the importance of each feature in a model’s prediction and have been used to interpret ML models in medical 60 and other domains 61,62 .”
Badal, V. D., Nebeker, C., Shinkawa, K., Yamada, Y., Rentscher, K. E., Kim, H., & Lee, E. (2021) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2021.728732 · Cited by 55
“A recent review on the topic, which details the advantages and major drawbacks can be found here (55).”
BackgroundPredicting Prolonged Length of ICU Stay through Machine Learning
Wu, J., Lin, Y., Li, P., Hu, Y., Zhang, L., & Kong, G. (2021) · Diagnostics · doi:10.3390/diagnostics11122242 · Cited by 62
“For most ML systems, an improved predictive accuracy may often be achieved through increased model complexity [66].”
BackgroundHuman-centered explainability for life sciences, healthcare, and medical informatics
Dey, S., Chakraborty, P., Kwon, B. C., Dhurandhar, A., Ghalwash, M. F., Suárez, F., Ng, K., Sow, D., Varshney, K. R., & Meyer, P. (2022) · Patterns · doi:10.1016/j.patter.2022.100493 · Cited by 72
“,(18,30,31) and Linardatos et al.(18,30,31) The first division separates whether the data or the model needs to be explained, although different personas might require both.”
BackgroundMethodInfluentialLaios, A., Kalampokis, E., Johnson, R. L., Munot, S., Thangavelu, A., Hutson, R., Broadhead, T., Theophilou, G., Leach, C., Nugent, D., et al. (2022) · Cancers · doi:10.3390/cancers14143447 · Cited by 51
Opening the black box: interpretable machine learning for predictor finding of metabolic syndrome
Zhang, Y., Zhang, X., Razbek, J., Li, D., Xia, W., Bao, L., Mao, H., Daken, M., & Cao, M. (2022) · BMC Endocrine Disorders · doi:10.1186/s12902-022-01121-4 · Cited by 51
“A crucial aspect of dividing the interpretability methods is based on the scale of interpretation, where local interpretability providesan explanation only for a specific instance, and global interpretability explains the whole model [26].”
Sargeant, H. (2022) · AI and Ethics · doi:10.1007/s43681-022-00236-7 · Cited by 76
Mukhamediev, R. I., Merembayev, T., Kuchin, Y. I., Malakhov, D. V., Zaitseva, E., Levashenko, V., Popova, Y., Сымагулов, А., Sagatdinova, G. N., & Amirgaliyev, Y. N. (2023) · Remote Sensing · doi:10.3390/rs15174269 · Cited by 55
Gupta, S., Kumar, A., & Maiti, J. (2024) · Safety Science · doi:10.1016/j.ssci.2024.106590 · Cited by 54
Maloney, K. O., Buchanan, C., Jepsen, R. D., Krause, K. P., Cashman, M. J., Gressler, B. P., Young, J. A., & Schmid, M. C. (2022) · Journal of Environmental Management · doi:10.1016/j.jenvman.2022.116068 · Cited by 49
Luo, Z., Zuo, R., Xiong, Y., & Zhou, B. (2023) · Natural Resources Research · doi:10.1007/s11053-023-10200-9 · Cited by 48
Graham, B., & Bonner, K. (2024) · Journal of Business Research · doi:10.1016/j.jbusres.2024.114567 · Cited by 49
“Recent advances in explainable artificial intelligence (XAI), such as the development of permutation importance and local interpretable model-agnostic explanations (LIME) have helped to overcome issues with model inter-pretability (Linardatos et al., 2021; Wang et al., 2022).”
BackgroundMiró-Nicolau, M., Jaume-i-Capó, A., & Moyà-Alcover, G. (2024) · Artificial Intelligence · doi:10.1016/j.artint.2024.104179 · Cited by 40
Thermodynamics-inspired explanations of artificial intelligence
Mehdi, S., & Tiwary, P. (2024) · Nature Communications · doi:10.1038/s41467-024-51970-x · Cited by 55
“Under the XAI paradigm, the developed methods can be black-box model-speci fi c, or model-agnostic thatgenerate globalor locally valid explanations in theform of visual or feature importance attributions 23 – 25 .”
A framework for anomaly classification in Industrial Internet of Things systems
Rodríguez, M., Tobón, D. P., & Múnera, D. (2024) · Internet of Things · doi:10.1016/j.iot.2024.101446 · Cited by 53
Artificial Intelligence for Monte Carlo Simulation in Medical Physics
Sarrut, D., Etxebeste, A., Muñoz, E., Krah, N., & Létang, J. M. (2021) · Frontiers in Physics · doi:10.3389/fphy.2021.738112 · Cited by 65
“A rich literature about machine learning interpretation methods exists [83], with a large part of the methods exploiting the gradient information fl owing through the layers of the network in order to highlight their impact.”
BackgroundAntimicrobial Resistance and Machine Learning: Challenges and Opportunities
Elyan, E., Hussain, A., Sheikh, A., Elmanama, A. A., Vuttipittayamongkol, P., & Hijazi, K. (2022) · IEEE Access · doi:10.1109/access.2022.3160213 · Cited by 48
Time to Assess Bias in Machine Learning Models for Credit Decisions
Brotcke, L. (2022) · Journal of risk and financial management · doi:10.3390/jrfm15040165 · Cited by 54
Deep Learning in Neuroimaging: Overcoming Challenges With Emerging Approaches
Smucny, J., Shi, G., & Davidson, I. A. K. (2022) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2022.912600 · Cited by 57
“In the context of brain imaging, they output individual heat maps, similar to statistical parametric maps outputted in voxelwise fMRI analysis, that illustrate the importance of particular voxels in their contribution to the decision for that scan (46).”
MethodGao, Z., & Mavris, D. N. (2022) · Aerospace · doi:10.3390/aerospace9120750 · Cited by 53
“A literature review of the interpretability methods in machine learning can be found in [123].”
Dawood, T., Chen, C., Sidhu, B. S., Ruijsink, B., Gould, J., Porter, B., Elliott, M. K., Mehta, V. S., Rinaldi, C. A., Anton, E. P., et al. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102861 · Cited by 65
“Specifically, the lack of trust in automated predictions for clinical applications is a major barrier preventing clinical adoption (Linardatos et al., 2021).”
BackgroundSynergi: A Mixed-Initiative System for Scholarly Synthesis and Sensemaking
Kang, H. B., Wu, T., Chang, J. C., & Kittur, A. (2023) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3586183.3606759 · Cited by 91
“[D4] Help scholars review the system-generated threads, curate ones that most interest them into their own outline, and iteratively build upon it.”
BackgroundExplainable Artificial Intelligence in Cybersecurity: A Brief Review
Hariharan, S., Velicheti, A., Anagha, A. S., Thomas, C., & Balakrishnan, N. (2021) · doi:10.1109/isea-isap54304.2021.9689765 · Cited by 54
Explainable inflation forecasts by machine learning models
Aras, S., & Lisboa, P. J. G. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117982 · Cited by 62
Kumar, S., & Das, A. (2023) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2023.107867 · Cited by 50
Mallio, C. A., Radbruch, A., Deike‐Hofmann, K., van der Molen, A. J., Dekkers, I. A., Zaharchuk, G., Parizel, P. M., Zobel, B. B., & Quattrocchi, C. C. (2023) · Investigative Radiology · doi:10.1097/rli.0000000000000983 · Cited by 53
Unraveling the landscape of large language models: a systematic review and future perspectives
Ding, Q., Ding, D., Wang, Y., Guan, C., & Ding, B. (2023) · Journal of Electronic Business & Digital Economics · doi:10.1108/jebde-08-2023-0015 · Cited by 58
A review: Insight into smart and sustainable ultra-precision machining augmented by intelligent IoT
Xu, Z., Zhu, T., Luo, F. L., Zhang, B., Poon, H., Yip, W. S., & To, S. (2024) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2024.03.008 · Cited by 50
Towards equitable AI in oncology
Viswanathan, V. S., Parmar, V. V., & Madabhushi, A. (2024) · Nature Reviews Clinical Oncology · doi:10.1038/s41571-024-00909-8 · Cited by 56
“The EU (European Union) has taken a stance against using deep learning for high-risk applications (which it broadly defines as AI relating to most private and public services, including health care), raising questions about the most effective regulatory pathway 160,161 .”
Haywood-Alexander, M., Liu, W., Bacsa, K., Lai, Z., & Chatzi, E. (2024) · Data-Centric Engineering · doi:10.1017/dce.2024.33 · Cited by 70
“In this case, it is imperative to deliver models that are explainable/interpretable and generalisable (Linardatos et al., 2020).”
Clinical deployment environments: Five pillars of translational machine learning for health
Harris, S. K., Bonnici, T., Keen, T., Lilaonitkul, W., White, M. J., & Swanepoel, N. (2022) · Frontiers in Digital Health · doi:10.3389/fdgth.2022.939292 · Cited by 54
“We argue that model explainability (Explainable Arti fi cial Intelligence [XAI]) methods need to be prioritized to help systematize and coordinate the processes of model troubleshooting by developers, risk-management by service providers, and system-checks by auditors (32 – 35).”
Towards explainable AI for hyperspectral image classification in Edge Computing environments
Lucia, G. D., Lapegna, M., & Romano, D. (2022) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2022.108381 · Cited by 49
Miralles, P., Thangavel, K., Scannapieco, A. F., Jagadam, N., Baranwal, P., Faldu, B., Abhang, R., Bhatia, S., Bonnart, S., Bhatnagar, I., et al. (2023) · Advances in Space Research · doi:10.1016/j.asr.2023.02.025 · Cited by 51
Potty, A. G. R., Potty, A. S. R., Maffulli, N., Blumenschein, L. A., Ganta, D., Mistovich, R. J., Fuentes, M., Denard, P. J., Sethi, P. M., Shah, A. A., et al. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12062369 · Cited by 47
The Past, the Present, and the Future of the Evolution of Mixed Reality in Teacher Education
Dieker, L., Hughes, C. E., & Hynes, M. C. (2023) · Education Sciences · doi:10.3390/educsci13111070 · Cited by 46
“This is not to say that LLMs cannot evolve to a level of transparency or as it often called, explain-able AI [57], in that the model can explain why it came to the conclusions it expressed.”
Zhao, F., Miao, F., Wu, Y., Ke, C., Gong, S., & Ding, Y. (2024) · Gondwana Research · doi:10.1016/j.gr.2024.02.011 · Cited by 43
Machine learning in industrial X-ray computed tomography – a review
Bellens, S., Guerrero, P., Vandewalle, P., & Dewulf, W. (2024) · CIRP journal of manufacturing science and technology · doi:10.1016/j.cirpj.2024.05.004 · Cited by 55
A machine learning approach for wind turbine power forecasting for maintenance planning
Dhungana, H. (2025) · Energy Informatics · doi:10.1186/s42162-024-00459-4 · Cited by 57
Artificial intelligence in variant calling: a review
Abdelwahab, O., & Torkamaneh, D. (2025) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2025.1574359 · Cited by 50
“Additionally, the interpretability of many AI models remains limited (Linardatos et al., 2020; Li X. et al., 2022), raising questions about reproducibility and user confidence, especially in regulatory and diagnostic workflows.”
Machine Learning in E-Commerce: Trends, Applications, and Future Challenges
Dritsas, E., & Τρίγκα, Μ. (2025) · IEEE Access · doi:10.1109/access.2025.3572865 · Cited by 62
“Businesses must balance predictive power with transparency to ensure accountability in automated decision-making [111], [112], [113], [114].”
Exploring Local Explanation of Practical Industrial AI Applications: A Systematic Literature Review
Le, T.-T.-H., Prihatno, A. T., Oktian, Y. E., Kang, H., & Kim, H. (2023) · Applied Sciences · doi:10.3390/app13095809 · Cited by 62
Raising awareness of sex and gender bias in artificial intelligence and health
Buslón, N., Cortés, À., Catuara‐Solarz, S., Cirillo, D., & Rementeria, M. J. (2023) · Frontiers in Global Women s Health · doi:10.3389/fgwh.2023.970312 · Cited by 55
A historical perspective of biomedical explainable AI research
Malinverno, L., Barros, V., Ghisoni, F., Visonà, G., Kern, R., Nickel, P. J., Ventura, B. E., Šimić, I., Stryeck, S., Manni, F., et al. (2023) · Patterns · doi:10.1016/j.patter.2023.100830 · Cited by 56
“The successes of foundation models, 57 which are those models trained on broad data and then adapted to specific tasks, have put in question whether any explainable model could reach the same level of performance.”
BackgroundGhani, S., Kumar, N., Gupta, M., & Saharan, S. (2023) · Asian Journal of Civil Engineering · doi:10.1007/s42107-023-00942-5 · Cited by 42
“However, these parameters exhibit intricate and non-linear interactions, making it challenging to derive a universally applicable equation that accurately encapsulates their correlations (Linardatos et al., 2020).”
Survey on ontology-based explainable AI in manufacturing
Naqvi, M. R., Elmhadhbi, L., Sarkar, A., Archimède, B., & Karray, M. H. (2024) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-023-02304-z · Cited by 49
“Nevertheless, such methods fall short of establishing a broad connection between both the predicted results and the source characteristics (Lipton, 2018a; Adadi & Berrada, 2018; Linardatos et al., 2020; Mohseni et al., 2018).”
Teshale, A. B., Htun, H. L., Vered, M., Owen, A. J., & Freak‐Poli, R. (2024) · Journal of Medical Systems · doi:10.1007/s10916-024-02087-7 · Cited by 51
“2 , these can be model-specific (use the structure of the model itself, e.g., built in feature importance measures in ensemble models) or model-agnostic (provide post-hoc explanations e.g., Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive exPlanations (SHAP)) [26].”
BackgroundExplainable artificial intelligence models in intrusion detection systems
Al, S., & Sağıroğlu, Ş. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110145 · Cited by 71
Vlontzou, M. E., Αθανασίου, Μ., Dalakleidi, K. V., Skampardoni, I., Davatzikos, C. A., & Nikita, K. S. (2025) · Scientific Reports · doi:10.1038/s41598-025-92577-6 · Cited by 52
“The continuously emerging need for fair, trustworthy, robust, and highly efficient ML models has led to the flourishing of eXplainable Artificial Intelligence (XAI), which entails the application of interpretability methods for producing explanations on the systems’ behaviour and final output 6 .”
Hossein‐Zadeh, N. G. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.110395 · Cited by 58
Tufariello, M., Pati, S., Palombi, L., Grieco, F., & Losito, I. (2022) · Foods · doi:10.3390/foods11070910 · Cited by 42
Assessing Human Motion During Exercise Using Machine Learning: A Literature Review
Frangoudes, F., Matsangidou, M., Schiza, E. C., Neokleous, K., & Pattichis, C. S. (2022) · IEEE Access · doi:10.1109/access.2022.3198935 · Cited by 57
“Interpretability, on the other hand, provides an understanding of why a specific outcome is observed and how it will change based on different inputs [145].”
BackgroundEnhancing Cluster Analysis With Explainable AI and Multidimensional Cluster Prototypes
Bobek, S., Kuk, M., Szelążek, M., & Nalepa, G. J. (2022) · IEEE Access · doi:10.1109/access.2022.3208957 · Cited by 43
“To avoid this issue, they propose an approximate column 241 generation algorithm that uses randomisation to efficiently 242 search the rule space and learn DNF or CNF classification 243 rules [16].”
MethodArtificial intelligence in oncologic imaging
Chen, M., Terzic, A., Becker, A. S., Johnson, J. M., Wu, C. C., Wintermark, M., Wald, C., & Wu, J. (2022) · European Journal of Radiology Open · doi:10.1016/j.ejro.2022.100441 · Cited by 46
“Interpretability of AI is another emerging area of focus in order to improve transparency and move away from the “black box” models [68].”
BackgroundAI in Computational Pathology of Cancer: Improving Diagnostic Workflows and Clinical Outcomes?
Çifçi, D., Veldhuizen, G. P., Foersch, S., & Kather, J. N. (2023) · Annual Review of Cancer Biology · doi:10.1146/annurev-cancerbio-061521-092038 · Cited by 54
Metadata
- Authors
- Linardatos, P., Papastefanopoulos, V., & Kotsiantis, S.
- Affiliations
- University of Patras (GR)
- Corresponding author
- Linardatos, Pantelis
- Venue
- Entropy
- Publisher
- MDPI AG
- Published
- 2020-12-25
- Volume
- 23
- Issue
- 1
- Pages
- 18
- ISSN
- 1099-4300 (ISSN-L)
- PMC
- PMC7824368
- OpenAlex
- W3116286104
- Cited by
- 2,922
- References
- 162
- Citations by year
- 2026: 485 · 2025: 702 · 2024: 701 · 2023: 565 · 2022: 370 · 2021: 97 · 2020: 1
- FWCI
- 8.89
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Entropy (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- 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 AI, model interpretability, black-box models, medical applications
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.