Neural Word Sense Disambiguation (WSD) has recently been shown to benefit from the incorporation of pre-existing knowledge, such as that coming from the WordNet graph. How- ever, state-of-the-art approaches have been successful in exploiting only the local struc- ture of the graph, with only close neighbors of a given synset influencing the prediction. In this work, we improve a classification model by recomputing logits as a function of both the vanilla independently produced logits and the global WordNet graph. We achieve this by incorporating an online neural approximated PageRank, which enables us to refine edge weights as well. This method exploits the global graph structure while keeping space requirements linear in the number of edges. We obtain strong improvements, matching the current state of the art. Code is available at https://github.com/SapienzaNLP/ neural-pagerank-wsd
Integrating personalized pagerank into neural word sense disambiguation / El Sheikh, Ahmed; Bevilacqua, Michele; Navigli, Roberto. - (2021), pp. 9092-9098. (Intervento presentato al convegno EMNLP 2021 tenutosi a Dominican Republic) [10.18653/v1/2021.emnlp-main.715].
Integrating personalized pagerank into neural word sense disambiguation
Michele BevilacquaSecondo
Conceptualization
;Roberto NavigliUltimo
Supervision
2021
Abstract
Neural Word Sense Disambiguation (WSD) has recently been shown to benefit from the incorporation of pre-existing knowledge, such as that coming from the WordNet graph. How- ever, state-of-the-art approaches have been successful in exploiting only the local struc- ture of the graph, with only close neighbors of a given synset influencing the prediction. In this work, we improve a classification model by recomputing logits as a function of both the vanilla independently produced logits and the global WordNet graph. We achieve this by incorporating an online neural approximated PageRank, which enables us to refine edge weights as well. This method exploits the global graph structure while keeping space requirements linear in the number of edges. We obtain strong improvements, matching the current state of the art. Code is available at https://github.com/SapienzaNLP/ neural-pagerank-wsdFile | Dimensione | Formato | |
---|---|---|---|
Navigli_integrating_2021.pdf
accesso aperto
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Creative commons
Dimensione
413.07 kB
Formato
Adobe PDF
|
413.07 kB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.