[1] 1. SECURE CLOUD COMPUTING WITH SECURE RESOURCES AND DATA COLOURING1.1 Trust and security are considered two main factors in cloud computing platform. Trust and security has prevented companies from fully embracing cloud platforms. Trust and reputation management has made cloud computing reliable with secure resources, data coloring and water marking. A reputation system is one of the ways to establish trust between service providers and data owners. These techniques are used to protect confidential data from unauthorized access. A new business model is enabled by cloud computing that supports on-demand, paid, and cost-effective IT services at Internet scale. The Internet cloud functions as a service factory built around virtualized data centers. Provision hardware, software, networks and datasets with virtualization to dynamically build cloud platforms. The basic idea is to migrate desktop computing to clusters of virtual servers and data centers. The ecological environment of the cloud should be secure, reliable and independent. You must first establish trust and security to increase the adoption of web and cloud services by cloud service providers (CPs). There should be no abuse, breach, hacking, cheating, rumors, viruses in a healthy cloud ecosystem. The main issue is establishing trust between cloud users, CSPs, and cloud data owners. A reputation-based trust management scheme, augmented with data coloring and software watermarking, is used to handle these issues. 1.2 Computing Trust Requirements in Cloud Services: Some Critical Issues for Trusted Cloud Computing and Several recent works discuss general cloud security issues and privacy has been identified by Cloud Computing......middle of paper.... .. and true, the probability that p4 is true is 0.8; in all other CASES THE PROBABILITY IS 0 (P4 IS FALSE). In corporate networks, graphical modes are important tools for analyzing security events. It may seem that the combination of Bayesian network and attack graph does not work well together, the real semantics and inference power of Bayesian networks are not fully utilized in this way. Bayesian networks are used very appropriately. To use Bayesian networks we must identify and correctly represent the relevant uncertainties. The Bayesian networks model is used to capture uncertain relationships, and the experimental results show that using Bayesian networks can provide new opportunities for better enterprise security analysis. To the best of our knowledge, our work is the first effort investigating systematic approaches for combining attack graphs and Bayesian networks