Abstract
This study introduces a method to assess the quality of Explainable Artificial Intelligence (XAI) algorithms in dynamic data streams, concentrating on the fidelity and stability of feature-importance and rule-based explanations. We employ XAI metrics, such as fidelity and Lipschitz Stability, to compare explainers between each other and introduce the Comparative Expert Stability Index (CESI) for benchmarking explainers against domain knowledge. We adopted the aforementioned metrics to the streaming data scenario and tested them in an unsupervised classification scenario with simulated distribution shifts as different classes. The necessity for adaptable explainers in complex scenarios, like failure detection is underscored, stressing the importance of continued research into versatile explanation techniques to enhance XAI system robustness and interpretability.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Alvarez-Melis, D., Jaakkola, T.S.: On the robustness of interpretability methods. In: Proceedings of the 35th International Conference on Machine Learning (ICML 2018), pp. 10–15. Stockholm, Sweden (2018)
Carrillo, G.A.: A model for designing rule-based expert systems (2017)
Cummins, L., et al.: Explainable predictive maintenance: a survey of current methods, challenges and opportunities (2024)
El-Khawaga, G., Elzeki, O., Abuelkheir, M., Reichert, M.: Why should i trust your explanation? an evaluation approach for XAI methods applied to predictive process monitoring results. IEEE Trans. Artif. Intell. PP, 1–15 (2024). https://doi.org/10.1109/TAI.2024.3357041
Elkhawaga, G., Elzeki, O., Abuelkheir, M., Reichert, M.: Evaluating explainable artificial intelligence methods based on feature elimination: a functionality-grounded approach. Electronics 12(7), 1670 (2023). https://doi.org/10.3390/electronics12071670
Hwang, G.H., Chen, B., Huang, S.H.: Development and analysis of an enhanced multi-expert knowledge integration system for designing context-aware ubiquitous learning contents. Int. J. Dist. Educ. Technol. 16(4), 1–16 (2018). https://doi.org/10.4018/IJDET.2018100103
Islam, M.R., Ahmed, M.U., Barua, S., Begum, S.: A systematic review of explainable artificial intelligence in terms of different application domains and tasks. Appl. Sci. 12(3) (2022). https://doi.org/10.3390/app12031353
Palli, A.S., et al.: Online machine learning from non-stationary data streams in the presence of concept drift and class imbalance: a systematic review. J. Inf. Commun. Technol. 23(1), 105–139 (2024). https://doi.org/10.32890/jict2024.23.1.5
Plazas, M., Ramos-Pollán, R., León, F., Martínez, F.: Towards reduction of expert bias on gleason score classification via a semi-supervised deep learning strategy. Proc. SPIE 12096, 120961I (2022). https://doi.org/10.1117/12.2611517
Rogowitz, B., Treinish, L.: How not to lie with visualization. Comput. Phys. 10 (1996). https://doi.org/10.1063/1.4822401
Rojat, T., Puget, R., Filliat, D., Ser, J.D., Gelin, R., Díaz-Rodríguez, N.: Explainable artificial intelligence (XAI) on timeseries data: a survey (2021)
Saarela, M., Geogieva, L.: Robustness, stability, and fidelity of explanations for a deep skin cancer classification model. Appl. Sci. 12(19) (2022). https://doi.org/10.3390/app12199545
Silva, M., Veloso, B., Gama, J.: Predictive maintenance, adversarial autoencoders and explainability, pp. 260–275 (2023). https://doi.org/10.1007/978-3-031-43430-3_16
Tavares, M., et al.: Expert knowledge integration in the data mining process with application to cardiovascular risk assessment. In: 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (2015). https://doi.org/10.1109/EMBC.2015.7318909
Veloso, B., Ribeiro, R.P., Gama, J., Pereira, P.M.: The metropt dataset for predictive maintenance. Sci. Data 9(1), 764 (2022)
Vergara-Lluri, M.E., et al.: Significant variability in the identification and reporting of band neutrophils by participants enrolled in the college of American pathologists proficiency testing program: time for a change. Archiv. Pathol. Lab. Med. 148(6), 666–676 (2023). https://doi.org/10.5858/arpa.2023-0015-CP
Vilone, G., Longo, L.: Notions of explainability and evaluation approaches for explainable artificial intelligence. Inf. Fusion 76, 89–106 (2021). https://doi.org/10.1016/j.inffus.2021.05.009
Acknowledgements
J. Gama and R. Ribeiro acknowledge the project AI-BOOST funded by the European Union under GA No 101135737. The paper is funded from the XPM project funded by the National Science Centre, Poland under the CHIST-ERA programme grant agreement Np. 857925 (NCN UMO-2020/02/Y/ST6/00070). The research has been supported by a grant from the Priority Research Area (DigiWorld) under the Strategic Programme Excellence Initiative at Jagiellonian University. We acknowledge the use of OpenAI’s ChatGPT-4 for reviewing and improving the language and style of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Ethics declarations
Disclosure of Interests
Apart from the funding mentioned in the Acknowledgements, the authors declare no competing interests.
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Mozolewski, M., Bobek, S., Ribeiro, R.P., Nalepa, G.J., Gama, J. (2024). Towards Evaluation of Explainable Artificial Intelligence in Streaming Data. In: Longo, L., Lapuschkin, S., Seifert, C. (eds) Explainable Artificial Intelligence. xAI 2024. Communications in Computer and Information Science, vol 2156. Springer, Cham. https://doi.org/10.1007/978-3-031-63803-9_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-63803-9_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-63802-2
Online ISBN: 978-3-031-63803-9
eBook Packages: Computer ScienceComputer Science (R0)