Automated Threat Attack Categorisation into Cloud Service Models

In today’s interconnected world, our heavy reliance on cloud- based services for daily tasks raises concerns about the security and privacy of our personal information. Cloud services have become an integral part of our lives, offering convenience and accessibility. However, the shared responsibility model and diverse service architectures of cloud computing necessitate different approaches to security assessment and penetration testing. This paper presents a comprehensive model-based approach that focuses on all three service models of the cloud. by adopting a system and model-based perspective, the study emphasizes on a shared approach to security. The methodology incorporates both white hat and grey hat techniques, which thoroughly test the security of cloud systems, ensuring a robust defense against various potential attacks. In addition to this, we are also present a novel approach for categorization using a combination of Long Short-Term Memory (LSTM) and Word2Vec, with a focus on Named Entity Recognition (NER). This approach achieves an impressive accuracy of 85 percent, which indicates its effectiveness in accurately identifying and categorizing named entities within the text data. The findings contribute in improving the overall security posture of cloud services, providing users with the confidence that their data is well protected.