OntoPedia: Automatic extraction of ontological and encyclopedic information about named entities

Design and application of techniques of natural language processing and information extraction in order to acquire, organize and maintain large amounts of information automatically encyclopedic. Specifically, this project addresses the creation of a system to classify and define entities with names, exploiting a corpus with encyclopedic knowledge constantly updated: Wikipedia and newspapers. This project works on three languages​​: Galician, Portuguese and Spanish.


  • Development of tools for dynamic construction of corpus.
  • Designing a dependency grammar and parsing of the corpus
  • Ranking of named entities.
  • Extraction of semantic relations from named entities.
  • Developed a base from encyclopedic knowledge and a information search system.
Link to the Project Website