A proposal of automatic selection of coarse-grained semantic classes for WSD

  1. Izquierdo-Bevia, Rubén
  2. Suárez Cueto, Armando
  3. Rigau Claramunt, Germán
Zeitschrift:
Procesamiento del lenguaje natural

ISSN: 1135-5948

Datum der Publikation: 2007

Nummer: 39

Seiten: 189-196

Art: Artikel

Andere Publikationen in: Procesamiento del lenguaje natural

Zusammenfassung

We present a very simple method for selecting Base Level Concepts using some basic structural properties of WordNet. We also empirically demonstrate that these automatically derived set of Base Level Concepts group senses into an adequate level of abstraction in order to perform class-based Word Sense Disambiguation. In fact, a very naive Most Frequent classifier using the classes selected is able to perform a semantic tagging with accuracy figures over 75%.