With the escalating number and complexity of cyber-attacks, in the realm of network security, intrusion detection systems (IDS) hold significant significance in protecting organizations against threats by continuously monitoring network traffic and analyzing patterns to swiftly identify and mitigate potential intrusions. This project aims to enhance the capabilities of an intrusion detection system by implementing a cloud-based architecture. By leveraging machine learning techniques, the system can adaptively learn from evolving attack patterns, improving its accuracy and proactive response to advanced threats. The motivation behind this project stems from traditional intrusion detection systems struggle to effectively combat the growing volume and complexity of cyber threats, it is essential to adapt and enhance our defense strategies, leaving organizations vulnerable to potential breaches and data compromises. The proposed hybrid intrusion detection system capitalizes on the advantages IDS techniques. It achieves this by leveraging signature-based detection to identify familiar patterns and employing anomaly-based detection to recognize unfamiliar threats, the model offers comprehensive coverage and improved accuracy in identifying and mitigating potential intrusions, enhancing overall system security. Experimental findings indicate that the system shows significant proficiency in identifying various classifications of network intrusion and achieves a high accuracy with minimal false positives compared to some of the recently proposed intrusion detection systems.