Literature Survey On Cloudsentry AI

Abstract

This paper presents a comprehensive survey of artificial intelligence–based intrusion detection and prevention systems (IDPS) for cloud environments, along with a proposed Transformer-based Spatio-Temporal Graph Neural Network (ST GNN) framework named Clouds entry AI. The survey analyzes existing methods including machine learning, deep learning, and hybrid intrusion detection approaches, highlighting their strengths and limitations. Identified gaps such as outdated datasets, lack of real-time validation, and high computational costs are addressed through the proposed ST-GNN model that learns spatial-temporal attack patterns efficiently. The study concludes that integrating Transformer attention with graph modeling can significantly enhance accuracy, scalability, and resilience for next-generation cloud security.