Improved Intelligent Malware Detection Model in Cloud Environment

Cloud computing security involves processes, policies, and technologies used by organizations and cloud providers to address threats and protect business data. These measures ensure data confidentiality, integrity, and availability within the cloud. However, the complex structure of the cloud requires significant security implementations to manage its challenges. Detecting malware in cloud environments is crucial for protecting against software attacks. This involves identifying and reducing the risk of malware, categorizing the cloud by severity levels, and enhancing the efficiency of incident response. Early classification methods are crucial, and our proposed methodology efficiently detects malware by selecting an appropriate Machine Learning (ML) model, improving its performance and evaluating it for enhanced detection capabilities, system performance, and removal of malware, spyware, and ransomware. To evaluate our approach, we utilized feature engineering, label encoding, and outlier removal techniques using Density-Based Spatial Clustering of Applications with Noise (DBSCAN). We applied K-Nearest Neighbor (K-NN), Random Forest Classifier (RFC), and Decision Tree [DT] algorithms, resulting in classification accuracy scores of 98, 98, and 99%.