A FRAMEWORK FOR EXPLAINABLE AI (XAI) IN MACHINE LEARNING-BASED FAKE NEWS DETECTION SYSTEMS: ENHANCING TRANSPARENCY, TRUST, AND USER AGENCY
DOI:
https://doi.org/10.32689/maup.it.2025.2.4Keywords:
Fake News Detection, Explainable AI (XAI), Machine Learning, Trustworthy AI, Conceptual Framework, Software Engineering, Misinformation, DisinformationAbstract
The subject of this article is the critical need for interpretability in machine learning (ML) based fake news detection systems and the proposal of a novel conceptual framework for the systematic integration of Explainable AI (XAI) to address this.The purpose is to enhance transparency, user trust, and effective moderation, thereby improving the fight against the significant threat of online disinformation.The proposed methodology involves delineating key architectural components for integrating XAI into fake news detection workflows, mapping diverse XAI techniques (e.g., LIME, SHAP, attention mechanisms) to the specific explainability needs of various stakeholders (end-users, journalists, moderators, developers), and considering the challenges of multimodal fake news. The potential benefits and operational characteristics of this framework are illustrated conceptually through mock experimental scenarios and illustrative case studies.The scientific novelty of this work lies in its comprehensive, stakeholder-centric XAI framework specifically tailored for the complexities of fake news detection. Unlike ad-hoc applications, this framework offers a systematic approach addressing multimodal content, outlining architectural considerations for integration, and linking explanation types to differentiated user requirements, aiming for a more holistic solution to the «black-box» problem in this domain.Conclusions from this conceptual study suggest that the proposed XAI framework provides a structured pathway towards developing more trustworthy, accountable, and effective AI-driven fake news detection systems. Its implementation is projected to enhance transparency, improve user agency in information assessment, facilitate model refinement, and support robust human-AI collaboration, thereby contributing a foundational approach for future empirical validation in combating online disinformation.
References
Adadi A., Berrada M. Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI). IEEE Access. 2018. Vol. 6. P. 52138–52160. DOI: 10.1109/ACCESS.2018.2870052.
Alaskar R., KP S. Explainable AI for deepfake detection: A review. MDPI Applied Sciences. 2023. Vol. 13, No. 5. Art. 3021. DOI: 10.3390/app13053021.
Arrieta A. B. et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion. 2020. Vol. 58. P. 82–115. DOI: 10.1016/j.inffus.2019.12.012.
Barredo Arrieta A. et al. On the explainability of mbox{Artifcial} Intelligence in fake news detection: mbox{Challenges} and future directions. TechRxiv. Preprint. 2022. DOI: 10.36227/techrxiv.19309205.v1.
Doshi-Velez F., Kim B. Towards A Rigorous Science of Interpretable Machine Learning. arXiv. Preprint arXiv:1702.08608. 2017. URL: https://arxiv.org/abs/1702.08608 (Last accessed: 17.05.2025).
Guidotti R. et al. A Survey of Methods for Explaining Black Box Models. ACM Computing Surveys. 2018. Vol. 51, No. 5. Art. 93. P. 1–42. DOI: 10.1145/3236009.
Linardatos P., Papastefanopoulos V., Kotsiantis S. Explainable AI: A Review of Machine Learning Interpretability Methods. Entropy. 2021. Vol. 23, No. 1. Art. 18. DOI: 10.3390/e23010018.
Lundberg S. M., Lee S.-I. A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems 30 (NIPS 2017). 2017. P. 4765–4774. URL: https://papers.nips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html (Last accessed: 17.05.2025).
Ribeiro M. T., Singh S., Guestrin C. "Why Should I Trust You?": Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). 2016. P. 1135–1144. DOI: 10.1145/2939672.2939778.
Shneiderman B. Bridging the gap between ethics and practice: guidelines for reliable, safe, and trustworthy Human-Centered AI systems. ACM Transactions on Interactive Intelligent Systems. 2020. Vol. 10, No. 4. Art. 26. P. 1–31. DOI: 10.1145/3419764.
Verma S., Dickerson J., Pruthi G. Counterfactual Explanations for Machine Learning: A Review. MLR : Workshop on Human Interpretability in Machine Learning (WHI 2020). 2020. URL: http://proceedings.mlr.press/v119/verma20a.html (Last accessed: 17.05.2025).
Zhang Y., Chen X. Explainable Recommendation: A Survey and New Perspectives. ACM Transactions on Intelligent Systems and Technology. 2020. Vol. 11, No. 5. Art. 50. P. 1–37. DOI: 10.1145/3383581.
