Neural sentiment analysis for a real-world application

Daniele Bonadiman, Giuseppe Castellucci, Andrea Favalli, Raniero Romagnoli, Alessandro Moschitti

Research output: Contribution to journalConference article

Abstract

In this paper, we describe our neural network models for a commercial application on sentiment analysis. Different from academic work, which is oriented towards complex networks for achieving a marginal improvement, real scenarios require flexible and efficient neural models. The possibility to use the same models on different domains and languages plays an important role in the selection of the most appropriate architecture. We found that a small modification of the state-of-the-art network according to academic benchmarks led to a flexible neural model that also preserves high accuracy.

Original languageEnglish
JournalCEUR Workshop Proceedings
Volume2006
Publication statusPublished - 1 Jan 2017

Fingerprint

Complex networks
Neural networks

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Bonadiman, D., Castellucci, G., Favalli, A., Romagnoli, R., & Moschitti, A. (2017). Neural sentiment analysis for a real-world application. CEUR Workshop Proceedings, 2006.

Neural sentiment analysis for a real-world application. / Bonadiman, Daniele; Castellucci, Giuseppe; Favalli, Andrea; Romagnoli, Raniero; Moschitti, Alessandro.

In: CEUR Workshop Proceedings, Vol. 2006, 01.01.2017.

Research output: Contribution to journalConference article

Bonadiman, D, Castellucci, G, Favalli, A, Romagnoli, R & Moschitti, A 2017, 'Neural sentiment analysis for a real-world application', CEUR Workshop Proceedings, vol. 2006.
Bonadiman D, Castellucci G, Favalli A, Romagnoli R, Moschitti A. Neural sentiment analysis for a real-world application. CEUR Workshop Proceedings. 2017 Jan 1;2006.
Bonadiman, Daniele ; Castellucci, Giuseppe ; Favalli, Andrea ; Romagnoli, Raniero ; Moschitti, Alessandro. / Neural sentiment analysis for a real-world application. In: CEUR Workshop Proceedings. 2017 ; Vol. 2006.
@article{181de50ed4dc4a0fbf9d0b7c02876663,
title = "Neural sentiment analysis for a real-world application",
abstract = "In this paper, we describe our neural network models for a commercial application on sentiment analysis. Different from academic work, which is oriented towards complex networks for achieving a marginal improvement, real scenarios require flexible and efficient neural models. The possibility to use the same models on different domains and languages plays an important role in the selection of the most appropriate architecture. We found that a small modification of the state-of-the-art network according to academic benchmarks led to a flexible neural model that also preserves high accuracy.",
author = "Daniele Bonadiman and Giuseppe Castellucci and Andrea Favalli and Raniero Romagnoli and Alessandro Moschitti",
year = "2017",
month = "1",
day = "1",
language = "English",
volume = "2006",
journal = "CEUR Workshop Proceedings",
issn = "1613-0073",
publisher = "CEUR-WS",

}

TY - JOUR

T1 - Neural sentiment analysis for a real-world application

AU - Bonadiman, Daniele

AU - Castellucci, Giuseppe

AU - Favalli, Andrea

AU - Romagnoli, Raniero

AU - Moschitti, Alessandro

PY - 2017/1/1

Y1 - 2017/1/1

N2 - In this paper, we describe our neural network models for a commercial application on sentiment analysis. Different from academic work, which is oriented towards complex networks for achieving a marginal improvement, real scenarios require flexible and efficient neural models. The possibility to use the same models on different domains and languages plays an important role in the selection of the most appropriate architecture. We found that a small modification of the state-of-the-art network according to academic benchmarks led to a flexible neural model that also preserves high accuracy.

AB - In this paper, we describe our neural network models for a commercial application on sentiment analysis. Different from academic work, which is oriented towards complex networks for achieving a marginal improvement, real scenarios require flexible and efficient neural models. The possibility to use the same models on different domains and languages plays an important role in the selection of the most appropriate architecture. We found that a small modification of the state-of-the-art network according to academic benchmarks led to a flexible neural model that also preserves high accuracy.

UR - http://www.scopus.com/inward/record.url?scp=85037328727&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037328727&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85037328727

VL - 2006

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

ER -