A Survey of Transformer-Based Stance Detection

View Sample PDF

Author(s): Dilek Küçük (TÜBİTAK Marmara Research Center, Turkey)
Copyright: 2023
Pages: 8
Source title: Deep Learning Research Applications for Natural Language Processing
Source Author(s)/Editor(s): L. Ashok Kumar (PSG College of Technology, India), Dhanaraj Karthika Renuka (PSG College of Technology, India)and S. Geetha (Vellore Institute of Technology, India)
DOI: 10.4018/978-1-6684-6001-6.ch004

Keywords: Artificial Intelligence / Engineering Science Reference / Knowledge Management and Library Science / Natural Language Processing

Purchase

View A Survey of Transformer-Based Stance Detection on the publisher's website for pricing and purchasing information.

Abstract

Stance detection systems are built in order to determine the position of text authors using the text that they produce and other contextual information. As the result of the stance detection procedure, the position of the text producer is determined as favor, against, or none. On the other hand, transformer-based technologies are reported to perform well for various natural language processing tasks. These are deep learning-based models that also incorporate attention mechanism. BERT and its variants are among the most popular transformer-based models proposed so far. In this chapter, the authors provide a plausible literature review on stance detection studies that are based on transformer models. Also included in the current chapter are important further research directions. Stance detection and transformer-based models are significant and recent problems in natural language processing and deep learning, respectively. Hence, they believe that this chapter will be an important guide for related researchers and practitioners working on these topics of high impact.

The IRMA Community

Research IRM

A Survey of Transformer-Based Stance Detection

Purchase

Abstract

Related Content

IRMA Sponsors