This paper presents sKGlable-VEO (scalable Knowledge Graph for Volcano Event detection), a framework designed for scalable seismic event detection through the integration of NeuroSymbolic AI and Knowledge Graphs. The system is structured around three modular pipelines: 1) a Knowledge Graph construction pipeline that transforms seismic data from Seismic Analysis Code (SAC) files into an ontology-based Knowledge Graph, 2) a Deep Learning training pipeline that trains neural network models on normalized seismic signals, and 3) an event detection pipeline that classifies seismic events using the trained models. Utilizing Docker containers, the sKGlable-VEO framework enables large-scale processing of seismic data while seamlessly integrating advanced AI models. This work advances seismic event detection by merging symbolic reasoning with machine learning in a scalable, efficient pipeline.
Scaling NeuroSymbolic AI Integration for Seismic Event Detection
Senatore, Sabrina;
2026
Abstract
This paper presents sKGlable-VEO (scalable Knowledge Graph for Volcano Event detection), a framework designed for scalable seismic event detection through the integration of NeuroSymbolic AI and Knowledge Graphs. The system is structured around three modular pipelines: 1) a Knowledge Graph construction pipeline that transforms seismic data from Seismic Analysis Code (SAC) files into an ontology-based Knowledge Graph, 2) a Deep Learning training pipeline that trains neural network models on normalized seismic signals, and 3) an event detection pipeline that classifies seismic events using the trained models. Utilizing Docker containers, the sKGlable-VEO framework enables large-scale processing of seismic data while seamlessly integrating advanced AI models. This work advances seismic event detection by merging symbolic reasoning with machine learning in a scalable, efficient pipeline.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
