Twitter as a Comparable Corpus to build Multilingual Affective Lexicons - Archive ouverte HAL Access content directly
Conference Papers Year :

Twitter as a Comparable Corpus to build Multilingual Affective Lexicons

Abstract

The main issue of any lexicon-based sentiment analysis system is the lack of affective lexicons. Such lexicons contain lists of words annotated with their affective classes. There exist some number of such resources but only for few languages and often for a small number of affective classes, generally restricted to two classes (positive and negative). In this paper we propose to use Twitter as a comparable corpus to generate a fine-grained and multilingual affective lexicons. Our approach is based in the co-occurence between English and target affective words in the same emotional corpus. And it can be applied to any number of target languages. In this paper we describe the building of affective lexicons for seven languages (en, fr, de, it, es, pt, ru).
Fichier principal
Vignette du fichier
BUCC14.pdf (523.3 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01615963 , version 1 (17-10-2017)

Identifiers

  • HAL Id : hal-01615963 , version 1

Cite

Amel Fraisse, Patrick Paroubek. Twitter as a Comparable Corpus to build Multilingual Affective Lexicons. The 7th Workshop on Building and Using Comparable Corpora, May 2014, Reykjavik, Iceland. pp.17-21. ⟨hal-01615963⟩
129 View
149 Download

Share

Gmail Facebook Twitter LinkedIn More