Problem and motivation
Traditional perimeter controls struggle with identity sprawl, short-lived workloads, and fragmented policy enforcement in multi-cloud serverless systems.
The research targets a missing capability: a serverless-native, AI-driven, end-to-end security architecture that can act across providers without losing zero-trust discipline.
Proposed architecture
SIF-CCA combines an XGBoost plus BiGRU detection pipeline with cloud-native event-driven orchestration and a unified zero-trust control plane.
Key reported metrics
- 98.00% accuracy, precision, recall, and F1-score.
- 0.990 ROC-AUC against the evaluated baselines.
- 135 ms average cross-cloud response latency.
- 99.6% policy consistency across providers.
Evidence from evaluation
Low off-diagonal noise indicates strong class separation across the paper’s target labels.
Baseline comparison
The proposed model keeps the strongest ROC envelope compared with RF, XGBoost, CNN, and LSTM.
Artifact access
The PDF and LaTeX sources are intentionally not exposed in plaintext on this public site. Use the secure gate to reveal the encrypted archives after completing the access policy steps.