Published Paper Details:

Security Operations Centers (SOCs) today face an overwhelming number of security alerts due to the rapid growth of cyber threats and widespread use of monitoring tools. While these alerts play a crucial role in identifying potential attacks, many of them turn out to be false positives or low-risk events. This often leads to alert fatigue, making it difficult for analysts to focus on real threats and respond efficiently. To address this issue, this project proposes an AI/ML-based alert prioritization system that intelligently filters and ranks security alerts from multiple sources such as SIEM systems, intrusion detection systems, firewalls, and endpoint security tools. The system processes alerts through stages like preprocessing, feature extraction, and supervised machine learning classification to identify high-risk incidents. A centralized dashboard displays prioritized alerts along with risk scores, helping analysts make faster and better decisions. By reducing unnecessary alert noise and improving visibility, the system enhances SOC performance, minimizes response time, and supports proactive threat detection in a secure and controlled environment.