Detecting Irony and Sarcasm in Twitter: The Role of Affective Content

There is a growing interest from the research community in investigating the impact of irony on sentiment analysis. Irony makes the life of a sentiment analysis tool quite difficult because in ironic opinions what is literally said is usually negated, and in absence of an explicit negation marker. A task has been organised in 2015 at SemEval on sentiment analysis of figurative language in Twitter. In this talk I will describe how irony is employed in tweets and reviews in general and what are the recent state-of-the-art attempts for its automatic detection. Linguistic devices such as ambiguity, incongruity, unexpectedness and emotional contexts play an important role as triggers of irony. In a recent work we proposed a novel model which explores the use of affective features based on a wide range of lexical resources. Classification experiments over different Twitter corpora show that affective information helps in distinguishing among ironic and non-ironic tweets. At the end I will also address the even more challenging fine-grained problem of discriminating between irony and sarcasm in order to study if #irony and #sarcasm can be considered as the same phenomenon in Twitter: #not.

Paolo Rosso is an associate professor of computer science at the Universitat Politècnica of València and is a member of the Pattern Recognition and Human Language Technology (PRHLT) research centre. His research interests include author profiling and irony detection in social media, opinion spam detection, as well as text reuse and plagiarism detection. Since 2009 he has been involved in the organisation of PAN benchmark activities, since 2010 and 2011 in the framework of CLEF and FIRE evaluation forums, on plagiarism / text reuse detection and author profiling. He has been also co-organiser of the shared tasks on sentiment polarity classification at Evalita and on sentiment analysis of figurative language in Twitter at SemEval-2015.