Explainable Artificial Intelligence: A Comprehensive Review of Methods, Applications, Challenges, and Future Research Directions
DOI:
https://doi.org/10.15680/IJCTECE.2025.0802016Keywords:
Explainable Artificial Intelligence, Explainable AI, LIME, SHAP, Grad-CAM, Human-Centered AIAbstract
Explainable Artificial Intelligence is an important area of research which addresses the lack of transparency of current artificial intelligence systems, particularly those utilizing deep learning and other sophisticated machine learning algorithms. This research paper presents an exhaustive review of Explainable Artificial Intelligence by discussing its past history, basic concepts, taxonomy, state-of-the-art explanation techniques, applications and research challenges. First, this article discusses the history of the development of AI from naturally interpretable machine learning algorithms to complex black-box deep learning architectures to highlight the increasing requirement of explainability. This article then goes ahead and categorizes the XAI techniques into intrinsic/post-hoc, local/global explanations and model-specific/model-agnostic techniques. In this research, we provide a critical review of popular explanation techniques including Local Interpretable Model-agnostic Explanations, Gradient-weighted Class Activation Mapping, Saliency Maps, Integrated Gradients, Attention Visualization, Counterfactual Explanations, Rule-based Explanation, Decision Tree and Feature Importance techniques in terms of their basic principles, advantages, disadvantages and applicability to various AI models.
References
[1] Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
[2] Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). Why should I trust you? Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135–1144).
[3] Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., & Torralba, A. (2016). Learning deep features for discriminative localization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 2921–2929).
[4] Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (Vol. 30, pp. 4765–4774).
[5] Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (pp. 618–626).
[6] Sundararajan, M., Taly, A., & Yan, Q. (2017). Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning (pp. 3319–3328).
[7] Samek, W., Wiegand, T., & Müller, K. R. (2017). Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models. arXiv preprint arXiv:1708.08296.
[8] Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
[9] Molnar, C. (2018). Interpretable Machine Learning. Lulu.com.
[10] Guidotti, R., Monreale, A., Ruggieri, S., Turini, F., Giannotti, F., & Pedreschi, D. (2018). A survey of methods for explaining black box models. ACM Computing Surveys, 51(5), 93:1–93:42.
