作者: Bernardo Magnini , Emanuele Pianta , Manuela Speranza , Octavian Popescu
DOI:
关键词:
摘要: In this paper we propose and investigate Ontology Population from Textual Mentions (OPTM), a sub-task of text where assume that mentions for several kinds entities (e.g. PERSON, ORGANIZATION , LOCATION GEOPOLITICAL_ ENTITY) are already extracted document collection. On the one hand, OPTM simplifies general task, limiting input textual material; on other it introduces challenging extensions to restricted named entities, being open wider spectrum linguistic phenomena. We describe manually created benchmark discuss factors which determine difficulty task.