Event Extraction and Semantic Representation from Spanish Workers' Statute Using Large Language Models

This work uses Large Language Models to process an important piece of Spanish legislation: the Workers' Statute. The proposed method extracts the relevant events in its articles using a GPT-3.5 model and represents the entities involved in the events and the relationships between them as RDF triples. The experiments carried out to select a high-performance strategy include both zero- and few-shot learning tests. Finally, this work proposes a strategy to uplift the extracted legal relations into a legal knowledge graph.