Learning a statistical model of product aspects for sentiment analysis
View/ Open
Impact
Scholar |
Other documents of the author: García Moya, Lisette; Berlanga Llavori, Rafael; Anaya Sánchez, Henry
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7038
comunitat-uji-handle3:10234/8634
comunitat-uji-handle4:
INVESTIGACIONMetadata
Title
Learning a statistical model of product aspects for sentiment analysisDate
2012Publisher
Sociedad Española para el Procesamiento del Lenguaje Natural (SEPLN)ISSN
1135-5948; 1989-7553Type
info:eu-repo/semantics/articleVersion
info:eu-repo/semantics/publishedVersionSubject
Abstract
En este art culo se introduce una nueva metodolog a para modelar ca-
racter sticas de productos a partir de una colecci on de opiniones de usuarios. La
metodolog a propuesta se basa en modelos estad sticos de ... [+]
En este art culo se introduce una nueva metodolog a para modelar ca-
racter sticas de productos a partir de una colecci on de opiniones de usuarios. La
metodolog a propuesta se basa en modelos estad sticos de lenguajes y es aplicable
a productos de dominio arbitrario. La metodolog a combina un kernel de palabras
de opini on con un modelo de traducci on de palabras para estimar el modelo de
caracter sticas. Se presenta adem as un m etodo para modelar las opiniones vertidas
sobre las caracter sticas. Los experimentos realizados sobre diferentes colecciones de
opiniones muestran resultados alentadores en el modelado tanto de caracter sticas
como de opiniones vertidas sobre estas [-]
In this paper, we introduce a new methodology for modeling product
aspects from a collection of free-text customer reviews. The proposal relies on a lan-
guage modeling framework and is domain independent. It combines ... [+]
In this paper, we introduce a new methodology for modeling product
aspects from a collection of free-text customer reviews. The proposal relies on a lan-
guage modeling framework and is domain independent. It combines both a kernel-
based model of opinion words and a stochastic translation model between words
to approach the aspect model of products. We also present a ranking-based met-
hodology to model the sentiments expressed about the aspects. The experiments
carried out over several collections of customer reviews show encouraging results in
the modeling of product aspects and their sentiments even from individual customer
reviews [-]
Is part of
Procesamiento del Lenguaje Natural, nº 49 septiembre de 2012Rights
http://rightsstatements.org/vocab/CNE/1.0/
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess
This item appears in the folowing collection(s)
- LSI_Articles [366]