SentinelSphere: AI-Driven Cybersecurity Platform Combining Threat Detection with Security Awareness

Abstract

The growing complexity of cyber threats coupled with the widening cybersecurity knowledge gap presents new challenges in the organisational security domain of the business sector. This paper introduces SentinelSphere, an innovative platform that redesigns cybersecurity defense by integrating advanced threat detection with interactive cybersecurity awareness education, creating a unified approach to building organisational cyber resilience. SentinelSphere employs an Enhanced Deep Neural Network model with specialised feature engineering to significantly reduce false positives while maintaining high detection accuracy across diverse attack vectors. The system includes a Traffic Light System (TLS) to transform complex threat intelligence into intuitive visual indicators, serving simultaneously as an operational tool for security professionals and an educational interface for non-technical users. This paper further presents a Large Language Model that delivers real-time, context-aware cybersecurity guidance and training. This conversational AI agent operates efficiently on standard enterprise hardware, making advanced security education accessible without requiring specialised infrastructure. SentinelSphere is validated using industry-standard datasets, achieving enterprise-grade performance in threat detection with a 94% F1 score and 69.5% reduction in false positives compared to baseline models. The system successfully processed nearly 11 million security events in 30 minutes, demonstrating scalability for enterprise deployment. This work contributes to the cybersecurity field by demonstrating that effective defense requires not just technological sophistication but also systematic enhancement of human security awareness.