Acronym: EXPRESS

Paula Cristina Quaresma da Fonseca Carvalho

From 01-Jul-2015

to 31-Dec-2016

Prime Contractor: INESC-ID (Other)

Financed by: FCT (Other)

Partnerships: INESC-ID (Other), University of Texas at Austin (University) - Austin, TX, USA;

Members: Paula Cristina Quaresma da Fonseca Carvalho, Bruno Emanuel da Graça Martins, Sílvio Amir Alves Moreira, Mário Jorge Costa Gaspar da Silva;

Abstract

The research carried out by EXPRESS will allow the production of:

(i) A thorough and innovative study on the expression of irony in text, taking into consideration different dimensions, namely: syntax, semantics, discourse, pragmatic context,domain and language.

(ii) A set of validated linguistic resources, namely fine-grained annotated corpora in Portuguese and in English describing irony.

(iii) Automated methods for irony detection, namely machine learning models, sensitive to target, domain and language. We believe that this project will help understanding the aspects related to the expression and recognition of irony in social media, both in Portuguese and in English, and will advance the state of the art in irony detection, in particular, and in subjective language processing (e.g sentiment analysis and opinion mining tasks) in general.