With its scalable, on-demand access to computer resources, cloud computing has completely transformed the IT landscape. Data availability, secrecy, and integrity are all jeopardized by the many security holes it opens up. In order to lessen the impact of cloud computing vulnerabilities, this study investigates the possibility of using machine learning (ML) method that is isolation forest. Anomaly detection, intrusion prevention, and risk assessment can all benefit from ML models’ pattern-detecting capabilities and large dataset analyses. Data reliance, computational complexity, and interpretability are some of the issues that ML encounters, despite the fact that it improves the accuracy and efficiency of vulnerability mitigation. We have shown that this method gave a 98.5 as an accuracy value in anomaly detection process that applied in the cloud dataset.