Sarcasm Detection using Context Separators in Online Discourse

Abstract

Sarcasm is an intricate form of speech, where meaning is conveyed implicitly. Being a convoluted form of expression, detecting sarcasm is an assiduous problem. The difficulty in recognition of sarcasm has many pitfalls, including misunderstandings in everyday communications, which leads us to an increasing focus on automated sarcasm detection. In the second edition of the Figurative Language Processing (FigLang 2020) workshop, the shared task of sarcasm detection released two datasets, containing responses along with their context sampled from Twitter and Reddit.In this work, we use RoBERT a large to detect sarcasm in both the datasets. We further assert the importance of context in improving the performance of contextual word embedding based models by using three different types of inputs -Response-only, Context-Response, and Context-Response (Separated). We show that our proposed architecture performs competitively for both the datasets. We also show that the addition of a separation token between context and target response results in an improvement of 5.13% in the F1-score in the Reddit dataset.

A Report on the 2020 Sarcasm Detection Shared Task

“…tanvidadu (Dadu and Pant, 2020): Fine-tuned RoBERTa-large model (355 Million parameters with over a 50K vocabulary size) on response and its two immediate contexts. They reported results on three different types of inputs: response-only model, concatenation of immediate two context with response, and using an explicit separator token between the response and the final context.…”

Sarcasm Detection: A Comparative Study

“…They report a performance with an accuracy of %79 on SARC dataset (Khodak et al, 2018). Similarly, Dadu and Pant (2020) use an ensemble of RoBERTa and ALBERT (Lan et al, 2019) on Get it #OffMyChest dataset (Jaidka et al, 2020) achieve a performance of %85 accuracy with F 1 score of 0.55. Javdan et al (2020) use BERT along with aspect-based sentiment analysis to extract the relation between context dialogue sequence and response.…”

Predicting Different Types of Subtle Toxicity in Unhealthy Online Conversations

“…Future work is also advised to test feature selection methods, which usually help increase classification accuracy by focusing on the most discriminating features while discarding redundant and irrelevant information. Lastly, given the challenging nature of unhealthy comments classification tasks, future work should look at specialized corpora (e.g., Wang and Potts, 2019;Oraby et al, 2017) and machine learning models trained at differentiating particular types of conversations, for example, solely sarcastic comments from non-sarcastic ones (e.g., Dadu and Pant, 2020;Hazarika et al, 2018).…”