Cloud-Based Data Intrusion Detection System Using Ml Techniques

Cloud computing (CC) stands as a groundbreaking technology, streamlining access to network and computer resources while offering a plethora of services, such as storage and data management, all with the aim of optimizing system functionality. Despite its array of advantages, cloud providers grapple with notable security challenges, particularly concerning the safeguarding of resources and services. Addressing these concerns and bolstering security measures necessitates vigilant monitoring of resources, services, and networks to promptly detect and respond to potential attacks. Central to this effort is the implementation of an advanced mechanism known as an intrusion detection system (IDS), which plays a pivotal role in regulating network traffic and identifying anomalous activities. This paper introduces an innovative cloud-based intrusion detection model that harnesses the power of the random forest (RF) algorithm alongside cutting-edge feature engineering techniques. Specifically, the integration of the RF classifier aims to enhance the accuracy (ACC) of the detection model significantly. The efficacy of the proposed model is rigorously evaluated using the NSL-KDD dataset, demonstrating a remarkable 99.99% ACC. This performance surpasses that of existing methodologies, underscoring the effectiveness and robustness of the proposed approach in addressing security challenges within cloud environments. Keywords —Accuracy, Anomaly detection, Cloud security, Feature Engineering, Intrusion Detection System (IDS), Random forest (RF)