Security Operation Centers (SOCs) are tasked with monitoring, detecting, and responding to a wide range of evolving cyber threats. However, current detection systems face major limitations including high false positive rates, slow response times, and poor adaptability to novel threats. Specifically, existing machine learning (ML) and deep learning (DL) approaches are often constrained by dataset imbalance, reliance on manual feature engineering, and lack of contextual understanding. There is a critical need for an integrated and intelligent threat detection system that can accurately classify and respond to diverse cyber threats such as: Web-based injection attacks (e.g., SQL Injection, Cross-Site Scripting), Spear-phishing emails, Trojan malware, Distributed Denial-of-Service (DDoS) attacks. Such a system must also reduce analyst workload by minimizing false alerts and automating reporting and remediation guidance.