Abstract
In the previous chapter, we presented approaches that capture incongruity within target text. However, as observed in errors reported by these approaches, some sarcastic text may require additional contextual information so that the sarcasm to be understood. This is true in case of sentences like ‘Nicki Minaj, don’t I hate her!’ or ‘Your parents must be really proud of you!’ These forms of sarcasm can be detected using contextual incongruity. Here, ‘contextual’ refers to information beyond the target text. In this chapter, we present approaches that capture contextual incongruity in order to detect sarcasm. We consider two settings. The first setting is a monologue (in Sect. 4.1) where a single author is being analyzed. In this case, we consider the historical context of the author, i.e., the text created by the author of the target text and create a sentiment map of entities. The second setting is a dialogue (in Sect. 4.2) where multiple participants take part in a conversation. In this case, we use sequence labeling as a novel formulation of sarcasm detection to capture contextual incongruity in the dialogue.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
- 2.
- 3.
Twitter API allows access to the most recent 3500 tweets on a timeline. This is an additional limitation.
- 4.
We also experimented with NN and JJ_NN sequences, but the output was unsatisfactory.
- 5.
- 6.
Some tweets in their original corpus could not be downloaded due to privacy settings or deletion.
- 7.
For example, some POS taggers have a separate tag for user mentions.
- 8.
- 9.
- 10.
- 11.
- 12.
For these scenes, the annotators later discussed and arrived at a consensus—they were then added to the dataset. The remaining scenes are done by either of the two annotators.
- 13.
The first utterance in a sequence has a null value for previous speaker.
- 14.
The two prior works are chosen based on what information was available in our dataset for the purpose of re-implementation. For example, approaches that use the Twitter profile information or the follower/friends structure in the Twitter, cannot be computed for our dataset.
- 15.
We also observe the same.
- 16.
- 17.
The work reports best accuracy of 65.44% for their dataset. This shows that our implementation is competent.
- 18.
The work reports best F-score of 67.8% for their dataset. This shows that our implementation is competent.
- 19.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Joshi, A., Bhattacharyya, P., Carman, M.J. (2018). Sarcasm Detection Using Contextual Incongruity. In: Investigations in Computational Sarcasm. Cognitive Systems Monographs, vol 37. Springer, Singapore. https://doi.org/10.1007/978-981-10-8396-9_4
Download citation
DOI: https://doi.org/10.1007/978-981-10-8396-9_4
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-8395-2
Online ISBN: 978-981-10-8396-9
eBook Packages: EngineeringEngineering (R0)