The DOI matches its Crossref record; 63 of the 64 references it cites were found in a registry or an index.
Shapley Values for Feature Selection: The Good, the Bad, and the Axioms
Open accessAbstract
The Shapley value has become popular in the Explainable AI (XAI) literature, thanks, to a large extent, to a solid theoretical foundation, including four “favourable and fair” axioms for attribution in transferable utility games. The Shapley value is provably the only solution concept satisfying these axioms. In this paper, we introduce the Shapley value and draw attention to its recent uses as a feature selection tool. We call into question this use of the Shapley value, using simple, abstract “toy” counterexamples to illustrate that the axioms may work against the goals of feature selection. From this, we develop a number of insights that are then investigated in concrete simulation settings, with a variety of Shapley value formulations, including SHapley Additive exPlanations (SHAP) and Shapley Additive Global importancE (SAGE). The aim is not to encourage any use of the Shapley value for feature selection, but we aim to clarify various limitations around their current use in the literature. In so doing, we hope to help demystify certain aspects of the Shapley value axioms that are viewed as “favourable”. In particular, we wish to highlight that the favourability of the axioms depends non-trivially on the way in which the Shapley value is appropriated in the XAI application.
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
“we incorporated the use of interpretability tools (eg, SHAP 28 values, feature importance) to ensure transparency, a priority emphasized in recent ML reporting guidelines.”
From Linguistic markers for identifying post-traumatic stress disorder and associated symptoms: a systematic literature review (2025) · cites this paper as [28]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“To align with best practices 27 for systematic reviews of ML studies, we augmented PROBAST with criteria tailored to these challenges”
From Linguistic markers for identifying post-traumatic stress disorder and associated symptoms: a systematic literature review (2025) · cites this paper as [28]
Experimental — accuracy not yet measured
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
Gauriat, C.-M., Pencolé, Y., Ribot, P., & Brouillet, G. (2024) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.4230/oasics.dx.2024.27 · Cited by 12,586
From local explanations to global understanding with explainable AI for trees
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S.-I. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-019-0138-9 · Cited by 10,551
17. A Value for n-Person Games
Shapley, L. S. (1953) · Contributions to the Theory of Games (AM-28), Volume II · doi:10.1515/9781400881970-018 · Cited by 6,950
A survey on feature selection methods
Chandrashekar, G., & Sahin, F. (2014) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2013.11.024 · Cited by 5,189
Shapley, L. S. (1988) · The Shapley Value · doi:10.1017/cbo9780511528446.003 · Cited by 4,954
Explaining prediction models and individual predictions with feature contributions
Štrumbelj, E., & Kononenko, I. (2014) · Knowledge and Information Systems · doi:10.1007/s10115-013-0679-x · Cited by 2,286
Tukey, J. W. (1962) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177704711 · Cited by 1,860
Consistent Individualized Feature Attribution for Tree Ensembles
Lundberg, S. M., Erion, G. G., & Lee, S.-I. (2018) · arXiv · doi:10.48550/arxiv.1802.03888 · Cited by 1,430
Budescu, D. V. (1993) · Psychological Bulletin · doi:10.1037/0033-2909.114.3.542 · Cited by 1,193
Research works printed but not linked: 1
- claeskens, 2008, Model Selection Model Averaging
Cited by
The Shapley Value in Machine Learning
Rozemberczki, B., Watson, L., Bayer, P., Yang, H.-T., Kiss, O., Nilsson, S., & Sarkar, R. (2022) · Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2022/778 · Cited by 296
The Shapley Value in Machine Learning
Rozemberczki, B., Watson, L., Bayer, P., Yang, H.-T., Kiss, O., Nilsson, S., & Sarkar, R. (2022) · arXiv · doi:10.48550/arxiv.2202.05594 · Cited by 312
“The feature selection game treats input features of a machine learning model as players and model performance as the payoff [Guyon et al. , 2003; Fryer et al. , 2021 Shapley values, and close relatives such as the Banzhaf index III, 1964 ] , have been studied as a measure of feature importance in…”
Methodvan Zyl, C., Ye, X., & Naidoo, R. (2024) · Applied Energy · doi:10.1016/j.apenergy.2023.122079 · Cited by 248
Nwakanma, C. I., Ahakonye, L. A. C., Njoku, J. N., Odirichukwu, J. C., Okolie, S. A., Uzondu, C., Ndubuisi Nweke, C. C., & Kim, D.-S. (2023) · Applied Sciences · doi:10.3390/app13031252 · Cited by 202
Explainable AI for Time Series Classification: A Review, Taxonomy and Research Directions
Theissler, A., Spinnato, F., Schlegel, U., & Guidotti, R. (2022) · IEEE Access · doi:10.1109/access.2022.3207765 · Cited by 194
On the failings of Shapley values for explainability
Huang, X., & Marques-Silva, J. (2024) · International Journal of Approximate Reasoning · doi:10.1016/j.ijar.2023.109112 · Cited by 165
A Review of Computational Modeling in Wastewater Treatment Processes
Duarte, M. S., Martins, G., Oliveira, P., Fernandes, B., Ferreira, E. C., Alves, M. M., Lopes, F., Pereira, M. A., & Nováis, P. (2023) · ACS ES&T Water · doi:10.1021/acsestwater.3c00117 · Cited by 184
Explainable AI for Safe and Trustworthy Autonomous Driving: A Systematic Review
Kuznietsov, A., Gyevnár, B., Wang, C., Peters, S., & Albrecht, S. V. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3474469 · Cited by 262
“Although this framework provides useful guidance, its application to AVs is not addressed and users’ queries are not considered.”
Brusa, E. G. M., Cibrario, L., Delprete, C., & Maggio, L. G. D. (2023) · Applied Sciences · doi:10.3390/app13042038 · Cited by 166
“In recent years, the SHAP value has become a popular method for the interpretation of the feature contributions in fitted ML models, as it can be seen in the references [47,52–57], posing the bases to Explainable Artificial Intelligence (XAI) [58] and new feature selection methods.”
MethodKibria, H. B., Nahiduzzaman, M., Goni, M. O. F., Ahsan, M., & Haider, J. (2022) · Sensors · doi:10.3390/s22197268 · Cited by 157
“The SHAP values have become very popular in explainable AI and are also used in feature selection [39], and model explanation [40].”
MethodEfficient Machine Learning Models for Early Stage Detection of Autism Spectrum Disorder
Bala, M., Ali, M. H., Satu, M. S., Hasan, K. F., & Moni, M. A. (2022) · Algorithms · doi:10.3390/a15050166 · Cited by 139
Aditya, P. S. R., & Pal, M. (2023) · International Journal of Data Science and Analytics · doi:10.1007/s41060-023-00458-w · Cited by 150
Designing a feature selection method based on explainable artificial intelligence
Zacharias, J., von Zahn, M., Chen, J., & Hinz, O. (2022) · Electronic Markets · doi:10.1007/s12525-022-00608-1 · Cited by 109
“There are various reasons for this task: The training and predictions of the model become computationally less expensive, the prediction power can be enhanced, and the interpretability increases (Guyon and Elisseeff, 2003; Fryer et al., 2021; Bhandari et al., 2020).”
BackgroundSurvey on Explainable AI: Techniques, challenges and open issues
Abusitta, A., Li, M. Q., & Fung, B. C. M. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124710 · Cited by 98
To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods
Amparore, E. G., Perotti, A., & Bajardi, P. (2021) · PeerJ Computer Science · doi:10.7717/peerj-cs.479 · Cited by 120
Multi-robot task allocation clustering based on game theory
Martín, J. G., Muros, F. J., Maestre, J. M., & Camacho, E. F. (2022) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2022.104314 · Cited by 99
Lin, P., Zhang, L., & Tiong, R. L. K. (2023) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2023.109172 · Cited by 91
Improving soil moisture prediction with deep learning and machine learning models
Teshome, F. T., Bayabil, H. K., Schaffer, B. A., Ampatzidis, Y. G., & Hoogenboom, G. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109414 · Cited by 107
Fu, H., Li, K., Zhang, C., Zhang, J., Liu, J., Chen, X., Chen, G., Sun, Y., Li, S., & Ling, L. (2023) · ACS Nano · doi:10.1021/acsnano.3c03610 · Cited by 74
Multimodal risk prediction with physiological signals, medical images and clinical notes
Wang, Y., Yin, C., & Zhang, P. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e26772 · Cited by 66
“Inspired by the explainable AI (XAI) works on feature importance explanation and model explanation [26–28], we calculated the Shapley value for the three modalities to show their contribution.”
MethodA review and benchmark of feature importance methods for neural networks
Mandler, H., & Weigand, B. (2024) · ACM Computing Surveys · doi:10.1145/3679012 · Cited by 75
“Fryer et al. [44], on the other hand, argued that the determining properties of Shapley values disqualify them for the use in feature selection tasks.”
Explainable and transparent artificial intelligence for public policymaking
Papadakis, T., Christou, I. T., Ipektsidis, C., Soldatos, J. K., & Amicone, A. (2024) · Data & Policy · doi:10.1017/dap.2024.3 · Cited by 71
“SHAP is based on the economic theory of the individual value a colleague brings to a collective effort of a finite number of individuals.”
MethodInfluentialAkbar, S., Ali, F., Hayat, M., Ahmad, A., Khan, S. A., & Gul, S. (2022) · Chemometrics and Intelligent Laboratory Systems · doi:10.1016/j.chemolab.2022.104682 · Cited by 70
Leveraging machine learning to proactively identify phishing campaigns before they strike
Zhang, K., Wang, H., Chen, M., Chen, X., Liu, L., Geng, Q., & Zhou, Y. C. (2025) · Journal Of Big Data · doi:10.1186/s40537-025-01174-x · Cited by 79
Atmakuru, A., Chakraborty, S., Faust, O., Salvi, M., Barua, P. D., Molinari, F., Acharya, U. R., & Homaira, N. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124665 · Cited by 66
Kanti, P. K., Paramasivam, P., Wanatasanappan, V. V., Dhanasekaran, S., & Sharma, P. (2024) · Scientific Reports · doi:10.1038/s41598-024-81955-1 · Cited by 63
“SHAP values for every feature would be computed following training so that the researchers may observe how factors such as temperature or concentration of nanoparticles affect the expected thermal characteristics 53,54 .”
Khan, S. A., Noor, S., Javed, T., Naseem, A., Aslam, F., AlQahtani, S. A., & Ahmad, N. (2025) · BioData Mining · doi:10.1186/s13040-024-00415-8 · Cited by 56
Koç, K., Ekmekcioğlu, Ö., & Gürgün, A. P. (2023) · Journal of Construction Engineering and Management · doi:10.1061/jcemd4.coeng-12848 · Cited by 67
Schiller, J., Stiller, S., & Ryo, M. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11165-2 · Cited by 77
“The grounding of SHAP in game theory is recognized as a fair and mathematically robust method (Fryer et al. 2021; Molnar 2023).”
Hettikankanamage, N. D., Shafiabady, N., Chatteur, F., Wu, M., Din, F. U., & Zhou, J. J. (2025) · Sensors · doi:10.3390/s25216649 · Cited by 68
Belfield, S. J., Cronin, M., Enoch, S. J., & Firman, J. W. (2023) · PLoS ONE · doi:10.1371/journal.pone.0282924 · Cited by 61
“Whilst it is clear that the level of insight offered by SHAP is considerable, it is nevertheless the case that methodological imperfections have been highlighted: some associated broadly with the application of Shapley values in the field of ML, and some concerning more the specifics of their implementation within the…”
BackgroundHeart Disease Prediction Using Stacking Model With Balancing Techniques and Dimensionality Reduction
Noor, A., Javaid, N., Alrajeh, N. A., Mansoor, B., Khaqan, A., & Bouk, S. H. (2023) · IEEE Access · doi:10.1109/access.2023.3325681 · Cited by 42
“This is especially beneficial for businesses and teams that report to clients or management [41].”
BackgroundŽivić, F., Kaplarevic-Malisic, A. M., Grujović, N., Stojanovic, B. S., & Ivanovic, M. R. (2025) · Materials Today Communications · doi:10.1016/j.mtcomm.2025.113525 · Cited by 65
Fan, M., Zhang, L., Liu, S., Yang, T., & Lu, D. (2023) · Frontiers in Water · doi:10.3389/frwa.2023.1112970 · Cited by 49
“It integrates several existing approaches with a solid theoretical foundation, guaranteeing that there is a unique solution for the measurement of additive feature importance scores (Lundberg and Lee, 2017; Sundararajan and Najmi, 2020; Fryer et al., 2021).”
BackgroundAkbar, S., Mohamed, H. G., Ali, H., Saeed, A., Ahmed, A., Gul, S., Ahmad, A., Ali, F., Ghadi, Y. Y., & Assam, M. (2023) · IEEE Access · doi:10.1109/access.2023.3274601 · Cited by 39
“The interpretable evaluation using a classification model also deals the issues that occur due to the lack of feature directivity [38].”
MethodDeep-ProBind: binding protein prediction with transformer-based deep learning model
Khan, S. A., Noor, S., Awan, H. H., Iqbal, S., AlQahtani, S. A., Dilshad, N., & Ahmad, N. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06101-8 · Cited by 39
“Decoding the biological import of selected features in machine learning models is some-times problematic since these algorithms are known as black boxes, and their internal workings are complex to understand [34].”
Noorchenarboo, M., & Grolinger, K. (2025) · Energy and Buildings · doi:10.1016/j.enbuild.2024.115177 · Cited by 38
“SHAP has gained considerable attention due to its solid theoretical foundation and its effectiveness in delivering reliable explanations across various domains, making it a widely accepted method for local explanations [19].”
Xu, L., Ning, S., Xu, X., Wang, S., Chen, L., Long, R., Zhang, S., Zhou, Y., Zhang, M., & Thapa, B. R. (2024) · Agricultural Water Management · doi:10.1016/j.agwat.2024.109176 · Cited by 43
What is the value of data? A review of empirical methods
Coyle, D., & Manley, A. (2023) · Journal of Economic Surveys · doi:10.1111/joes.12585 · Cited by 35
Nguyen, V. G., Sharma, P., Ağbulut, Ü., Le, H. S., Cao, D. N., Dzida, M., Osman, S. M., Le, H. C., & Trần, V. D. (2024) · International Journal of Green Energy · doi:10.1080/15435075.2024.2326076 · Cited by 42
“Shapley values, which are derived from the theory of cooperative games, are used to divide “credit” for prediction among the qualities of inputs in a manner that is both fair and consistent (Fryer, Strumke, and Nguyen 2021; Wojtuch, Jankowski, and Podlewska 2021).”
Xu, T., Tian, A., Gao, J., Yan, H., & Liu, C. (2024) · Environmental Modelling & Software · doi:10.1016/j.envsoft.2024.106194 · Cited by 37
Gene selection with Game Shapley Harris hawks optimizer for cancer classification
Afreen, S., Bhurjee, A. K., & Aziz, R. M. (2023) · Chemometrics and Intelligent Laboratory Systems · doi:10.1016/j.chemolab.2023.104989 · Cited by 33
Wu, X., Li, L., Tao, X., Yuan, J., & Xie, H. (2025) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2025.3537918 · Cited by 38
“Therefore, we retain the top k happiness factors and exclude the others to obtain a concise explanation [73], which is preferable for human decisions over one that includes all factors. c) Fidelity: It measures how well Shapley value approximates the behavior of the black-box models for the sample…”
Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
Bhangu, S. K., Welch, N., Lewis, M., Li, F., Gardner, B., Thissen, H. W., & Kowalczyk, W. (2025) · Small Science · doi:10.1002/smsc.202400579 · Cited by 37
Akbar, S., Ali, H., Ahmad, A., Sarker, M. R., Saeed, A., Salwana, E., Gul, S., Khan, A., & Ali, F. (2023) · IEEE Access · doi:10.1109/access.2023.3268523 · Cited by 31
FCE: Feedback Based Counterfactual Explanations for Explainable AI
Suffian, M., Graziani, P., Alonso, J. M., & Bogliolo, A. (2022) · IEEE Access · doi:10.1109/access.2022.3189432 · Cited by 36
Liu, J., Zhang, Q., & Macián‐Juan, R. (2024) · Progress in Nuclear Energy · doi:10.1016/j.pnucene.2024.105287 · Cited by 32
Nguyen, V. G., Paramasivam, P., Dzida, M., Osman, S. M., Le, D. T. N., Cao, D. N., Truong, T. H., & Trần, V. D. (2024) · Case Studies in Thermal Engineering · doi:10.1016/j.csite.2024.104743 · Cited by 32
Efremov, C., Le, T. T., Paramasivam, P., Rudzki, K., Osman, S. M., & Chau, T. H. (2024) · International Journal of Hydrogen Energy · doi:10.1016/j.ijhydene.2024.11.329 · Cited by 29
Afreen, S., Bhurjee, A. K., & Aziz, R. M. (2024) · Microchemical Journal · doi:10.1016/j.microc.2024.111280 · Cited by 29
Petelin, G., Cenikj, G., & Eftimov, T. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.119023 · Cited by 32
Zhou, H., Yue, X., Wang, B., Tian, C., Lu, X., Zhu, J., & Cao, Y. (2023) · Environmental Research Letters · doi:10.1088/1748-9326/ad059c · Cited by 29
“This method directly regulates the effects of each feature on model losses or gains using the differences between predicted and expected values (Fryer et al 2021).”
MethodPhysics-informed deep learning framework for explainable remaining useful life prediction
Kim, M., Yoo, S., Son, S., Chang, S. Y., & Oh, K. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110072 · Cited by 31
Kallah-Dagadu, G., Mohammed, M., Nasejje, J. B., Mchunu, N. N., Twabi, H. S., Batidzirai, J. M., Singini, G. C., Nevhungoni, P., & Maposa, I. (2025) · Scientific Reports · doi:10.1038/s41598-025-85323-5 · Cited by 35
“However, there are criticisms to the shapely values, as highlighted in a study by 61,62 , cautioning against their use.”
Al-Helali, B. M., Chen, Q., Xue, B., & Zhang, M. (2024) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3369407 · Cited by 31
A comparative study of methods for estimating model-agnostic Shapley value explanations
Olsen, L. H. B., Glad, I. K., Jullum, M., & Aas, K. (2024) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-024-01016-z · Cited by 57
Li, F., Jin, J., Segers, A. J., Lin, H. X., Pang, M., Xiao, C., Deng, T., & Liao, H. (2022) · Geoscientific model development · doi:10.5194/gmd-15-7791-2022 · Cited by 27
Orthogonal least squares based fast feature selection for linear classification
Zhang, S., & Lang, Z. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108419 · Cited by 25
“Recently, it has been developed as a unified approach to the feature importance analysis for machine learning models [38], which can also be used for feature selection [39].”
MethodKoç, K., Budayan, C., Ekmekcioğlu, Ö., & Tokdemir, O. B. (2023) · Journal of Construction Engineering and Management · doi:10.1061/jcemd4.coeng-13857 · Cited by 33
Constable, P. A., Pinzon-Arenas, J. O., Díaz, L. R. M., Lee, I., Marmolejo‐Ramos, F., Loh, L., Zhdanov, A. E., Kulyabin, M., Brabec, M., Skuse, D., et al. (2024) · Bioengineering · doi:10.3390/bioengineering12010015 · Cited by 25
“Another feature selection approach is the calculation of Shapley values [51,52], which, as FI , measure each features’ impact on the output of the ML model [53].”
Identifying Student Profiles Within Online Judge Systems Using Explainable Artificial Intelligence
Rico-Juan, J. R., Sánchez-Cartagena, V. M., Valero-Mas, J. J., & Gallego, A. (2023) · IEEE Transactions on Learning Technologies · doi:10.1109/tlt.2023.3239110 · Cited by 28
“More precisely, we resort to the so-called Shapley Additive Explanations (SHAP) introduced in [44], as it represents one of the most commonly considered techniques for XAI in the literature [45].”
MethodA New Multi-Class Rebalancing Framework for Imbalance Medical Data
Edward, J., Rosli, M. M., & Seman, A. (2023) · IEEE Access · doi:10.1109/access.2023.3309732 · Cited by 26
“features and has shown to be a reliable alternative feature selection approach [38].”
MethodExplainable ensemble models for predicting wall thickness loss of water pipes
Taiwo, R., Yussif, A., Seghier, M. E. A. B., & Zayed, T. M. (2024) · Ain Shams Engineering Journal · doi:10.1016/j.asej.2024.102630 · Cited by 26
An Interpretable Electric Vehicles Battery State of Charge Estimation Using MHDTCN-GRU
K, A. P. N., Rithish, J., Nath, A. G., Singh, S. K., & Singh, R. K. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3447228 · Cited by 21
Advances in Explainable Artificial Intelligence (xAI) in Finance
Klein, T., & Walther, T. (2024) · Finance research letters · doi:10.1016/j.frl.2024.106358 · Cited by 27
Predicting Patient Length of Stay in Australian Emergency Departments Using Data Mining
Gurazada, S. G., Gao, S., Burstein, F. V., & Buntine, P. (2022) · Sensors · doi:10.3390/s22134968 · Cited by 24
“This will help to determine predictive accuracy of more advanced models when compared to DM [74–76].”
ResultThe benefits and dangers of using machine learning to support making legal predictions
Zeleznikow, J. (2023) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.1505 · Cited by 34
“Such examples include Lahav et al. (2018), Fryer et al. (2021), Kovalerchuk et al. (2021) and Watson (2022).”
BackgroundTransferring Black-Box Decision Making to a White-Box Model
Žlahtič, B., Završnik, J., Vošner, H. B., & Kokol, P. (2024) · Electronics · doi:10.3390/electronics13101895 · Cited by 26
Paramasivam, P., Alruqi, M., & Ağbulut, Ü. (2024) · Fuel · doi:10.1016/j.fuel.2024.133093 · Cited by 19
Petropoulos, A., Stavroulakis, E., Lazaris, P., Siakoulis, V., & Vlachogiannakis, N. E. (2023) · Working paper - Bank of Greece · doi:10.52903/wp2023315 · Cited by 20
Koç, K. (2023) · Journal of Management in Engineering · doi:10.1061/jmenea.meeng-5492 · Cited by 24
A machine learning‐based exploration of resilience and food security
Villacis, A. H., Badruddoza, S., & Mishra, A. K. (2024) · Applied Economic Perspectives and Policy · doi:10.1002/aepp.13475 · Cited by 28
Shapley Value as an Aid to Biomedical Machine Learning: a Heart Disease Dataset Analysis
Scapin, D., Cisotto, G., Gindullina, E., & Badia, L. (2022) · IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid) · doi:10.1109/ccgrid54584.2022.00113 · Cited by 21
Explainable machine learning to predict the cost of capital
Bussmann, N., Giudici, P., Tanda, A., & Yu, E. P. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1578190 · Cited by 23
ShapFuzz: Efficient Fuzzing via Shapley-Guided Byte Selection
Zhang, K., Zhu, X., Xiao, X., Xue, M., Zhang, C., & Wen, S. (2024) · Network and Distributed System Security Symposium · doi:10.14722/ndss.2024.23134 · Cited by 17
“Moreover, the Shapley value has been applied in numerous fields, such as quantifying the importance of a feature in deep learning models [11].”
MethodApplications and Computation of the Shapley Value in Databases and Machine Learning
Luo, X., & Pei, J. (2024) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3626246.3654680 · Cited by 25
“Firstly, by modeling features as players, the Shapley value is applied in feature selection [12, 23] and model interpretation [15, 35, 36], in order to quantify the contribution of individual features to the overall model performance.”
MethodBreast Cancer Prediction Using Shapely and Game Theory in Federated Learning Environment
Supriya, Y., & Chengoden, R. (2024) · IEEE Access · doi:10.1109/access.2024.3424934 · Cited by 22
“When referring to the survey related to Shapley values, the authors in [36] use Shapley values for feature selection.”
MethodLu, Y., Xu, C., & Wang, X. (2025) · Ecological Indicators · doi:10.1016/j.ecolind.2025.113711 · Cited by 16
Importance of 3D convolution and physics on a deep learning coastal fog model
Kamangir, H., Krell, E., Collins, W. G., King, S. A., & Tissot, P. E. (2022) · Environmental Modelling & Software · doi:10.1016/j.envsoft.2022.105424 · Cited by 16
“Shapley values have been proposed to add rigor to XAI (Messalas et al. (2019); Fryer et al. (2021)).”
BackgroundMulti-view graph network learning framework for identification of major depressive disorder
Zhang, M., Long, D., Chen, Z., Fang, C., Li, Y., Huang, P., Chen, F., & Sun, H. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107478 · Cited by 22
Predicting dynamic stability from static features in power grid models using machine learning
Titz, M., Kaiser, F. J., Kruse, J., & Witthaut, D. (2024) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/5.0175372 · Cited by 23
Raufi, B., & Longo, L. (2024) · BioMedInformatics · doi:10.3390/biomedinformatics4010048 · Cited by 19
“The Shapley values quantify the contribution of each feature to the model’s performance on a given set of data points [44].”
Angelucci, G., Quaranta, G., Mollaioli, F., & Kunnath, S. K. (2024) · Journal of Building Engineering · doi:10.1016/j.jobe.2024.110124 · Cited by 17
“Apart from being useful to understand how machine learning models work, SHAP interpretation has been also proposed to assist the selection of the most relevant explanatory variables involved into a model [47].”
Dubey, Y. K., Tarte, Y., Talatule, N., Damahe, K., Palsodkar, P., & Fulzele, P. R. (2024) · Diagnostics · doi:10.3390/diagnostics14222514 · Cited by 17
“Understanding their strengths and limits might assist practitioners in selecting the best approach for their interpreting needs [28,29].”
Stress management with HRV following AI, semantic ontology, genetic algorithm and tree explainer
Chatterjee, A., Riegler, M. A., Ganesh, K., & Halvorsen, P. (2025) · Scientific Reports · doi:10.1038/s41598-025-87510-w · Cited by 19
AI Opacity and Explainability in Tort Litigation
Fraser, H., Simcock, R., & Snoswell, A. J. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533084 · Cited by 15
“Examples of Local XAI methods include Shapley values and SHAP force and dependence graphs [30, 51], LIME [61], and scoped rules or ‘Anchors’ [60].”
MethodZhang, L., Li, C., Zhang, G., Wu, X., Zhou, L., Chen, L., Jiao, Y., Liu, G., & Hei, W. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.111014 · Cited by 16
Francis, D. P., Laustsen, M., Dossi, E., Treiberg, T., Hardy, I., Shiv, S. H., Hansen, B. S., Mogensen, J., Jakobsen, M. H., & Alstrøm, T. S. (2023) · Analytical Methods · doi:10.1039/d3ay00247k · Cited by 14
Terahertz Time-Domain Spectroscopy of Glioma Patient Blood Plasma: Diagnosis and Treatment
Черкасова, О. П., Vrazhnov, D. A., Knyazkova, A. I., Konnikova, M. R., Stupak, E. V., Glotov, V., Stupak, V. V., Nikolaev, N. A., Георгиевич, П. А., Peng, Y., et al. (2023) · Applied Sciences · doi:10.3390/app13095434 · Cited by 14
FedIMP: Parameter Importance-based Model Poisoning attack against Federated learning system
Li, X., Wang, N., Yuan, S., & Guan, Z. (2024) · Computers & Security · doi:10.1016/j.cose.2024.103936 · Cited by 18
Ye, Z., Li, K., Chen, G., Zou, Y., Wang, S., Wang, Z., Yu, S., Wang, Z., Jia, D., & Zhou, Y. (2024) · Corrosion Science · doi:10.1016/j.corsci.2024.112388 · Cited by 14
An Explainable AI based Clinical Assistance Model for Identifying Patients with the Onset of Sepsis
Chakraborty, S., Kumar, K., Pailla, B. R., Meena, T., & Roy, S. S. (2023) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri58017.2023.00059 · Cited by 13
“Deep SHAP makes use of Deep Learning Important FeaTures (DeepLIFT) [22] and Shapley value [23] and assigns each feature an importance value for a particular prediction.”
MethodChakraborty, S., Kumar, K., Tadepalli, K., Pailla, B. R., & Roy, S. S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17828-y · Cited by 13
“Deep SHAP is the combination of Deep Learning Important FeaTures (DeepLIFT) [37, 38] and Shapley value [39].”
SVCE: Shapley Value Guided Counterfactual Explanation for Machine Learning-Based Autonomous Driving
Li, M., Sun, H., Huang, Y., & Chen, H. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3393634 · Cited by 14
“On the one hand, Shapley values can measure the feature contribution of models developed on tabular data [3], [20], [21], which helps us understand the impact of each feature on the overall model performance.”
Online monitoring of Haematococcus lacustris cell cycle using machine and deep learning techniques
Stegemüller, L., Caccavale, F., Valverde‐Pérez, B., & Angelidaki, İ. (2024) · Bioresource Technology · doi:10.1016/j.biortech.2024.131976 · Cited by 14
Fasniuc-Pereu, E., Drăgoi, E.-N., Bulgariu, D., Popescu, M., & Bulgariu, L. (2025) · Polymers · doi:10.3390/polym17152089 · Cited by 12
“If the first cause is quite unlikely (but not impossible), to check the second possible cause, the SHapley Additive exPlanations (SHAP) values [39] were computed using the SHAP module.”
The effects of urban green infrastructures on the housing market
Liu, X., Zhu, L., Wang, Z., Zhang, X., Zhang, X., Chen, Z., & Qiao, R. (2025) · Habitat International · doi:10.1016/j.habitatint.2025.103629 · Cited by 11
Hu, B., Xu, Y., Gong, H., Tang, L., & Li, H. (2024) · Academic Radiology · doi:10.1016/j.acra.2024.08.040 · Cited by 11
Vlasov, V. V., Kuzin, A. Y., Florya, I., Buriak, N. S., Chernyshev, V. S., Golikov, A. D., Krasnov, L. V., Mikhaylov, S., Pugach, M. A., Nasibulin, A. G., et al. (2025) · Journal of Energy Storage · doi:10.1016/j.est.2025.115349 · Cited by 13
Žlahtič, B., Završnik, J., Vošner, H. B., Kokol, P., Šuran, D., & Završnik, T. (2023) · Applied Sciences · doi:10.3390/app13148329 · Cited by 12
Metadata
- Authors
- Fryer, D., Strumke, I., & Nguyen, H.
- Affiliations
- The University of Queensland (AU)
- La Trobe University (AU)
- Venue
- IEEE Access
- Published
- 2021
- Volume
- 9
- Pages
- 144352-144360
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3131037858
- Semantic Scholar
- 6573fa74af238d1e3538c026997e31b9f67f19f7
- Cited by
- 347
- References
- 64
- Citations by year
- 2026: 56 · 2025: 102 · 2024: 86 · 2023: 58 · 2022: 34 · 2021: 11
- FWCI
- 16.63
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- IEEE Access (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- arXiv (Cornell University)
- La Trobe University (cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Game Theory and Voting Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Keywords
- Shapley value, feature selection, explainable AI, axioms, SHAP, sage, counterexample, transferable utility games, attribution
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.