Aproximación a la categorización textual en español basada en la semántica de marcos
ISSN: 1135-5948
Argitalpen urtea: 2008
Zenbakia: 41
Orrialdeak: 65-71
Mota: Artikulua
Beste argitalpen batzuk: Procesamiento del lenguaje natural
Laburpena
FrameNet is a resource based on Frame Semantics that comprises how languages account for daily situations linguistically. Frames represent information packets about how to convey information about a certain situation. This paper presents an approach to categorize texts by analysing the range of FrameNet situations that co-occur in a particular text. The set of FrameNet situations is used as a feature vector where the presence or absence of certain frames in a text is used to determine its category. Results show how our system was able to categorize texts in Spanish with high accuracy.