Measuring the quality of lexical-semantic resources is a challenging problem. In this paper, we describe a general approach for quality evaluation in lexical-semantic resources in terms of the quality of their synsets. We also introduce a complete definition for the quality of lexical-semantic resources as a set of synset incorrectness, incompleteness, and connectivity measures that evaluate all synset components. This study demonstrates that synset quality is a summation process that integrates the quality measures of synset components. Furthermore, we then address the main challenges that affect the optimal quality achievement of lexical-semantic resources. Our work, thus, serves to evaluate the quality of monolingual and multilingual lexical-semantic resources and achieves accurate results in natural language processing (NLP) applications.
The Dimensions of Lexical Semantic Resource Quality / Khalilia, Hadi; Freihat, Abed Alhakim; Giunchiglia, Fausto. - (2021), pp. 38-44. (Intervento presentato al convegno 2nd International Workshop on NLP Solutions for Under Resourced Languages, NSURL 2021 tenutosi a Trento nel 12th - 13th November 2021).
The Dimensions of Lexical Semantic Resource Quality
Khalilia, Hadi;Freihat, Abed Alhakim;Giunchiglia, Fausto
2021-01-01
Abstract
Measuring the quality of lexical-semantic resources is a challenging problem. In this paper, we describe a general approach for quality evaluation in lexical-semantic resources in terms of the quality of their synsets. We also introduce a complete definition for the quality of lexical-semantic resources as a set of synset incorrectness, incompleteness, and connectivity measures that evaluate all synset components. This study demonstrates that synset quality is a summation process that integrates the quality measures of synset components. Furthermore, we then address the main challenges that affect the optimal quality achievement of lexical-semantic resources. Our work, thus, serves to evaluate the quality of monolingual and multilingual lexical-semantic resources and achieves accurate results in natural language processing (NLP) applications.File | Dimensione | Formato | |
---|---|---|---|
2021.nsurl-1.3.pdf
accesso aperto
Tipologia:
Post-print referato (Refereed author’s manuscript)
Licenza:
Creative commons
Dimensione
173.52 kB
Formato
Adobe PDF
|
173.52 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione