Increasing the Effectiveness of the Romanian Wordnet in NLP Applications
Închide
Articolul precedent
Articolul urmator
632 0
SM ISO690:2012
BARBUMITITELU, Verginica. Increasing the Effectiveness of the Romanian Wordnet in NLP Applications. In: International Conference on Intelligent Information Systems, 22-23 august 2013, Chișinău. Chișinău: "VALINEX" SRL, 2013, pp. 122-125.
EXPORT metadate:
Google Scholar
Crossref
CERIF

DataCite
Dublin Core
International Conference on Intelligent Information Systems 2013
Conferința "International Conference on Intelligent Information Systems"
Chișinău, Moldova, 22-23 august 2013

Increasing the Effectiveness of the Romanian Wordnet in NLP Applications


Pag. 122-125

BarbuMititelu Verginica
 
Institute for Artificial Intelligence, Romanian Academy
 
 
Disponibil în IBN: 12 februarie 2018


Rezumat

The Romanian wordnet is a valuable linguistic resource: it contains 52357 literals distributed in 58725 synsets and covers 86175 senses, which makes it one of the richest such resources worldwide. Besides its intrinsic value, it is also used in various NLP applications developed in our group. In this paper, we present a scenario that proves that enriching the Romanian wordnet with derivational relations increases its effectiveness in Question Answering tasks. The statement is true for whatever wordnet, irrespective of the language it represents.

Cuvinte-cheie
wordnet,

derivational relations, lexical chain