Questo paper mira a colmare il gap di accuratezza tra il constituency parsing dell’Italiano e quello Inglese: come primo miglioramento, abbiamo adattato il parser a costituenti per l’Inglese, Bllip, anche noto come Charniak parser, per l’Italiano e lo abbiamo addestrato sul Turin University Treebank. In seguito, abbiamo progettato un reranker basato sulle Macchine a Vettori di Supporto che usano kernel arborei, i quali possono efficacemente generalizzare pattern sintattici, richiedendo pochi dati di training per addestrare il modello. Il nostro approccio supera lo stato dell’arte ottenuto con il Berkeley parser, migliorando la labeled F1 da 84.54 a 86.81.

This paper aims at filling the gap between the accuracy of Italian and English constituency parsing: firstly, we adapt the Bllip parser, i.e., the most accurate constituency parser for English, also known as Charniak parser, for Italian and trained it on the Turin University Treebank (TUT). Secondly, we design a parse reranker based on Support Vector Machines using tree kernels, where the latter can effectively generalize syntactic patterns, requiring little training data for training the model. We show that our approach outperforms the state of the art achieved by the Berkeley parser, improving it from 84.54 to 86.81 in labeled F1

Tree Kernels-based Discriminative Reranker for Italian Constituency Parsers / Uva, Antonio; Moschitti, Alessandro. - ELETTRONICO. - 2:(2016), pp. 303-307. [10.4000/books.aaccademia.1860]

Tree Kernels-based Discriminative Reranker for Italian Constituency Parsers

Antonio Uva;Alessandro Moschitti
2016

Abstract

This paper aims at filling the gap between the accuracy of Italian and English constituency parsing: firstly, we adapt the Bllip parser, i.e., the most accurate constituency parser for English, also known as Charniak parser, for Italian and trained it on the Turin University Treebank (TUT). Secondly, we design a parse reranker based on Support Vector Machines using tree kernels, where the latter can effectively generalize syntactic patterns, requiring little training data for training the model. We show that our approach outperforms the state of the art achieved by the Berkeley parser, improving it from 84.54 to 86.81 in labeled F1
Proceedings of the Third Italian Conference on Computational Linguistics' (CLiC–it 2016)
Torino
Accademia University Press
9788899982546
Uva, Antonio; Moschitti, Alessandro
Tree Kernels-based Discriminative Reranker for Italian Constituency Parsers / Uva, Antonio; Moschitti, Alessandro. - ELETTRONICO. - 2:(2016), pp. 303-307. [10.4000/books.aaccademia.1860]
File in questo prodotto:
File Dimensione Formato  
Parse_Rerank-IT.pdf

accesso aperto

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Creative commons
Dimensione 233.86 kB
Formato Adobe PDF
233.86 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11572/169258
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact