The present research exploits the largeamount of linguistic resources developedinto the Lexicon-grammar paradigm in thedomain of the Opinion Mining. Groundedon the Semantic Predicates theory, theproposed system is able to automaticallymatch the syntactic structures selected byspecial classes of verbs, indicating positiveor negative Sentiment, Opinion or Physical acts, with the semantic frames evokedby the same lexical items. This methodshas been tested on a large dataset com-posed of short texts, such as tweets andnews headings.

Towards a lexicon-grammar based framework for NLP: an opinion mining application

Pelosi Serena;Guarasci Raffaele
2015

Abstract

The present research exploits the largeamount of linguistic resources developedinto the Lexicon-grammar paradigm in thedomain of the Opinion Mining. Groundedon the Semantic Predicates theory, theproposed system is able to automaticallymatch the syntactic structures selected byspecial classes of verbs, indicating positiveor negative Sentiment, Opinion or Physical acts, with the semantic frames evokedby the same lexical items. This methodshas been tested on a large dataset com-posed of short texts, such as tweets andnews headings.
2015
Istituto di Calcolo e Reti ad Alte Prestazioni - ICAR
Inglese
Mitkov, Ruslan and Angelova, Galia and Bontcheva, Kalina
Proceedings of the International Conference Recent Advances in Natural Language Processing
International Conference Recent Advances in Natural Language Processing
160
167
8
https://www.aclweb.org/anthology/R15-1023
INCOMA Ltd.
Shoumen
BULGARIA
Sì, ma tipo non specificato
5-11/09/2015
Hissar, Bulgaria
Natural Language Processing
Lexicon-Grammar
Opinion Mining
4
open
Elia, Annibale; Pelosi, Serena; Maisto, Alessandro; Guarasci, Raffaele
273
info:eu-repo/semantics/conferenceObject
04 Contributo in convegno::04.01 Contributo in Atti di convegno
File in questo prodotto:
File Dimensione Formato  
R15-1023.pdf

accesso aperto

Licenza: Creative commons
Dimensione 106.03 kB
Formato Adobe PDF
106.03 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: https://hdl.handle.net/20.500.14243/419667
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? ND
social impact