In this paper we present the results of the evaluation of an automatic mapping between two lexical resources, WordNet/ItalWordNet and ImagAct, a conceptual ontology of action types instantiated by video scenes. Results are compared with those obtained from a previous experiment performed only on Italian data. Differences between the two evaluation strategies, as well as between the quality of the mappings for the two languages considered in this paper, are iscussed.
L'articolo presenta i risultati della valutazione di un mapping automatico realizzato tra due risorse lessicali, WordNet/ItalWordNet e ImagAct, un'ontologia concettuale di tipi azionali rappresentati per mezzo di video. Tali risultati vengono confrontati con quelli ottenuti da un precedente esperimento, condotto esclusivamente sull'italiano. Vengono inoltre discusse le differenze tra le due strategie di valutazione, cos`? come nella qualit`a del mapping proposto per le due lingue qui considerate.
Evaluating ImagAct-WordNet mapping for English and Italian through videos
De Felice I;Bartolini R;Russo I;Quochi V;Monachini M
2014
Abstract
In this paper we present the results of the evaluation of an automatic mapping between two lexical resources, WordNet/ItalWordNet and ImagAct, a conceptual ontology of action types instantiated by video scenes. Results are compared with those obtained from a previous experiment performed only on Italian data. Differences between the two evaluation strategies, as well as between the quality of the mappings for the two languages considered in this paper, are iscussed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
