This research introduces a novel Three-in-One (TIO) System that unifies violence detection, retrieval, and explanation through knowledge graphs and Graph Attention Networks.

• Integrates multimodal analysis with knowledge graphs for enhanced violence detection
• Utilizes Graph Attention Networks (GAT) to improve classification accuracy
• Provides explainable results instead of black-box predictions
• Enables unified detection, retrieval, and explanation in a single system

For security applications, this approach offers significant advantages by providing transparent, interpretable threat assessments that security professionals can effectively evaluate and trust, moving beyond opaque AI decisions in critical safety scenarios.

Detection, Retrieval, and Explanation Unified: A Violence Detection System Based on Knowledge Graphs and GAT