Multilingual Extraction Ontologies

Abstract The growth of multilingual web content and increasing internationaliza-tion portends the need for cross-language query processing. We offer ML-OntoES (a MultiLingual Ontology-based Extraction System) as a solution for narrow-domain/data-rich applications. Based on language-independent extraction ontolo-gies (Embley, Liddle, & Lonsdale, 2011), ML-OntoES enables semantic search over domain-specific, semi-structured information. Key ideas of ML-OntoES include: (1) monolingual semantic indexing and query interpretation with extraction ontologies and (2) conceptual-level cross-language translation. A prototype implementation, along with experimental work showing good extraction accuracy in multiple lan-guages, demonstrates the viability of the ML-OntoES approach of using multilin-gual extraction ontologies for cross-language query processing. Key words: Cross-language query processing, extraction ontologies, monolingual semantic indexing, monolingual query interpretation, conceptual-level cross-lan-guage information transfer.