This project is funded by Ministerio de Ciencia e Innovación.
The objective of this project is to develop a Question-Answering system by extracting ontology and encyclopedic information of Named Entities (NE), using Wikipedia as external resource. Our work is focused on the following topics: question-answering, named entity recognition and classification, semantic relation extraction, open information extraction, and dependency-based parsing.
Within this project, we have designed ONTOpedia, a search engine that retrieves basic enciclopedic facts (triplets) from a large knowledge base (47GB), automatically extracted from four language versions of Wikipedia: English, Portuguese, Spanish, and Galician.
Information Extraction Dependency Grammar Open Information Extraction Named Entity Classification Parsing
ProLNat@GE is a research team belonging to Grupo de Gramática do Español, Universidade de Santiago de Compostela