EUVIMEDEUVIMED entdecken
Crossref

Explainable AI Models for Secure and Transparent Decision-Making in Smart Computing Environments

Shallu Yadav

International Journal of Emerging Technologies and Innovative Research · 30.06.2026 · 64

DOI: 10.48175/ijarsct-6608 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Emerging Technologies and Innovative Research, online erschienen am 30.06.2026. Verfasst von Shallu Yadav. Die Arbeit nennt 40 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid proliferation of smart computing environments—encompassing the Internet of Things (IoT), edge intelligence, cyber-physical systems, and autonomous control loops—has made automated decision-making indispensable for security, safety, and operational efficiency. However, the high-performing machine learning and deep learning models that drive these decisions typically behave as opaque “black boxes,” offering little insight into why a particular outcome was produced. In security-critical settings such as intrusion detection, access control, and anomaly response, this opacity erodes trust, complicates auditing, and impedes regulatory compliance. This paper proposes an Explainable Artificial Intelligence (XAI) framework that couples a hybrid CNN-LSTM classifier with a post-hoc explanation layer based on SHAP, LIME, and attention weighting to deliver secure and transparent decisions. The framework was evaluated on a benchmark network-security dataset comprising normal and attack traffic across five categories. The proposed model attained an accuracy of 97.8%, an F1-score of 97.4%, and an area under the ROC curve (AUC) of 0.991, outperforming Random Forest, Support Vector Machine, XGBoost, and Multilayer Perceptron baselines. Beyond predictive accuracy, the explanation layer achieved a fidelity of 0.94 and a trust score of 0.89, while exposing the dominant decision drivers (e.g., packet inter-arrival time, flow byte rate, and failed authentication counts) in human-readable form. The results demonstrate that explainability and security are mutually reinforcing rather than competing objectives, and that transparent rationale generation can be embedded into smart computing pipelines with negligible loss of predictive power. The study contributes a reproducible methodology, a layered reference architecture, and a multi-dimensional evaluation protocol that jointly measures accuracy, robustness, and explanation quality

Zitieren

APA

Yadav, S. (2026). Explainable AI Models for Secure and Transparent Decision-Making in Smart Computing Environments. International Journal of Emerging Technologies and Innovative Research, 64. https://doi.org/10.48175/ijarsct-6608

Vancouver

Yadav S. Explainable AI Models for Secure and Transparent Decision-Making in Smart Computing Environments. International Journal of Emerging Technologies and Innovative Research. 2026:64. doi:10.48175/ijarsct-6608

BibTeX
@article{Yadav2026,
  author = {Yadav, Shallu},
  title = {Explainable AI Models for Secure and Transparent Decision-Making in Smart Computing Environments},
  journal = {International Journal of Emerging Technologies and Innovative Research},
  year = {2026},
  pages = {64},
  doi = {10.48175/ijarsct-6608}
}
Kostenlos starten →