DWAEF: A Deep Weighted Average Ensemble Framework Harnessing Novel Indicators for Sarcasm Detection

Tracking #: 740-1720


Responsible editor: 

Jodi Schneider

Submission Type: 

Research Paper

Abstract: 

Sarcasm is a linguistic phenomenon often indicating a disparity between literal and inferred meanings. Due to its complexity, it is typically difficult to discern it within an online text message. Consequently, in recent years sarcasm detection has received considerable attention from both academia and industry. Nevertheless, the majority of current approaches simply model low-level indicators of sarcasm in various machine learning algorithms. This paper aims to present sarcasm in a new light by utilizing novel indicators in a Deep Weighted Average Ensemble-based Framework (DWAEF). The novel indicators pertain to exploiting the presence of simile and metaphor in text and detecting the subtle shift in tone at a sentence's structural level. A Graph Neural Network (GNN) structure is implemented to detect the presence of simile, Bidirectional Encoder Representations from Transformers (BERT) embeddings are exploited to detect metaphorical instances and Fuzzy Logic is employed to account for the shift of tone. To account for the existence of sarcasm, the DWAEF integrates the inputs from the novel indicators. The performance of the framework is evaluated on a self-curated dataset of online text messages. A comparative report between the results acquired using conventional features and those obtained using proposed indicators is provided. The encouraging findings produced after applying DWAEF demonstrate that the proposed method surpasses the outcomes of previous research that made use of primitive features.

Manuscript: 

Tags: 

  • Under Review

Data repository URLs: 

Self curated dataset was used for the present research work

(is made available to reviewers and will be published upon acceptance)

Date of Submission: 

Saturday, December 31, 2022