Real World Ontology

Real-world ontology research focuses on creating and utilizing knowledge representations that accurately model the complexities of real-world domains, aiming to improve AI systems' understanding and interaction with these domains. Current efforts concentrate on developing methods for continual learning of evolving ontologies, leveraging large language models to capture everyday knowledge, and improving the semantic richness of machine learning models through techniques like contextual embeddings and lattice-preserving embeddings of description logics. This research is significant for advancing AI capabilities in areas like autonomous driving and personal assistants, as well as for improving the efficiency and scalability of ontology engineering and knowledge graph construction.