Sentiment Analysis on Monolingual, Multilingual and Code-Switching Twitter Corpora

Abstract

[Abstract]: We address the problem of performing po- larity classification on Twitter over differ- ent languages, focusing on English and Spanish, comparing three techniques: (1) a monolingual model which knows the language in which the opinion is written, (2) a monolingual model that acts based on the decision provided by a language iden- tification tool and (3) a multilingual model trained on a multilingual dataset that does not need any language recognition step. Results show that multilingual models are even able to outperform the monolingual models on some monolingual sets. We introduce the first code-switching corpus with sentiment labels, showing the robust- ness of a multilingual approach.