The role of idioms in sentiment analysis
In this project we investigate the role of idioms in automated approaches to sentiment analysis.
First, to support idioms as features we collected a set of 580 idioms that are relevant to sentiment analysis, i.e. the ones that can be mapped to a sentiment. These mappings were then obtained using a web-based crowdsourcing approach. The quality of the crowdsourced information is demonstrated with high agreement among five independent annotators calculated using Krippendorff's alpha coefficient (α = 0.662).
Second, to evaluate the results of sentiment analysis, we assembled a corpus of sentences in which idioms are used in context. Each sentence was annotated with a sentiment.
The results of sentiment analysis experiments are described in the paper below.
Idioms annotated with sentiment polarity.
Sentences that contain idioms, also annotated with sentiment polarity.
Local grammars that model idioms from the list above, which have been implemented in Mixup language. Matched idioms can be viewed using MinorThird (see screenshot above). An example of how to run this Mixup file can be found here.
Lowri Williams, Christian Bannister, Michael Arribas Ayllon, Alun Preece and Irena Spasić (2014) The role of idioms in sentiment analysis, submitted