5Towards a Multilingual System for Automatic Irony Detection
5.1. Introduction
In Chapters 3 and 4, we proposed a multilevel annotation scheme for irony and an automatic irony detection system for tweets in French. In this chapter, we shall assess the portability of both the annotation scheme and the automatic detection system for other languages. Two experiments will be presented.
The first experiment consists of testing the portability of the multilevel annotation scheme for irony, described in Chapter 3, for other languages in the same family as French: English and Italian. An annotation campaign is carried out using our scheme, designed to analyze pragmatic phenomena relating to irony in French tweets. The aim of this first experiment is not only to test the performance of the annotation scheme on other Indo-European languages that are culturally similar to French, but also to measure the impact of a set of pragmatic phenomena on the interpretation of irony. Additionally, we shall consider the way in which these phenomena interact with the local context of a tweet in languages belonging to the same family.
The second experiment consists of testing the performance of the feature-based automatic irony detection system (using the SurfSystem and PragSystem models – see Chapter 4) on tweets written in Arabic. For the purposes of this experiment, we constructed the first corpus of ironic and non-ironic tweets in Arabic, studying the performance of features and assessing the ...
Get Automatic Detection of Irony now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
