Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Computational Approaches to the Detection of Lesser-Known Rhetorical Figures: A Systematic Survey and Research Challenges

Ramona Kühn, Jelena Mitrović, Michael Granitzer

TL;DR

The paper addresses the problem of computationally detecting lesser-known rhetorical figures, which are often overlooked by NLP systems. It surveys 26 figures across 39 papers, categorizing detection methods into rule-based and model-based approaches, and highlights pervasive issues such as dataset scarcity and English-language bias. A key contribution is the aggregation of datasets, definitions, and detection strategies, along with a critical discussion of methodological gaps and the potential of ontologies (e.g., RhetFig, GRhOOT) and multilingual resources. The work underscores the practical importance of recognizing rhetorical devices for improving tasks like hate speech detection, sentiment analysis, and humor understanding, and it advocates for richer datasets, multilingual research, and integration of rhetorical ontologies and LLMs to advance the field.

Abstract

Rhetorical figures play a major role in our everyday communication as they make text more interesting, more memorable, or more persuasive. Therefore, it is important to computationally detect rhetorical figures to fully understand the meaning of a text. We provide a comprehensive overview of computational approaches to lesser-known rhetorical figures. We explore the linguistic and computational perspectives on rhetorical figures, emphasizing their significance for the domain of Natural Language Processing. We present different figures in detail, delving into datasets, definitions, rhetorical functions, and detection approaches. We identified challenges such as dataset scarcity, language limitations, and reliance on rule-based methods.

Computational Approaches to the Detection of Lesser-Known Rhetorical Figures: A Systematic Survey and Research Challenges

TL;DR

The paper addresses the problem of computationally detecting lesser-known rhetorical figures, which are often overlooked by NLP systems. It surveys 26 figures across 39 papers, categorizing detection methods into rule-based and model-based approaches, and highlights pervasive issues such as dataset scarcity and English-language bias. A key contribution is the aggregation of datasets, definitions, and detection strategies, along with a critical discussion of methodological gaps and the potential of ontologies (e.g., RhetFig, GRhOOT) and multilingual resources. The work underscores the practical importance of recognizing rhetorical devices for improving tasks like hate speech detection, sentiment analysis, and humor understanding, and it advocates for richer datasets, multilingual research, and integration of rhetorical ontologies and LLMs to advance the field.

Abstract

Rhetorical figures play a major role in our everyday communication as they make text more interesting, more memorable, or more persuasive. Therefore, it is important to computationally detect rhetorical figures to fully understand the meaning of a text. We provide a comprehensive overview of computational approaches to lesser-known rhetorical figures. We explore the linguistic and computational perspectives on rhetorical figures, emphasizing their significance for the domain of Natural Language Processing. We present different figures in detail, delving into datasets, definitions, rhetorical functions, and detection approaches. We identified challenges such as dataset scarcity, language limitations, and reliance on rule-based methods.

Paper Structure

This paper contains 115 sections, 3 figures, 17 tables.

Table of Contents

  1. Introduction
  2. Contributions and Structure of the Paper
  3. The Linguistic Perspective on Rhetoric and Rhetorical Figures
  4. The Computational Perspective on Rhetorical Figures
  5. Computational Detection of Rhetorical Figures
  6. Alliteration
  7. Definition
  8. Related Figures
  9. Rhetorical Effect
  10. Domain
  11. Datasets
  12. Detection Approach
  13. Figures with Perfect Lexical Repetition
  14. Definitions
  15. Related Figures
  16. ...and 100 more sections

Figures (3)

  • Figure 1: The structure of this survey. We cover the linguistic background briefly. Our main focus lies on the formalization of rhetorical figures, annotated datasets, and the detection of various figures. We are not considering the computational interpretation of rhetorical figures.
  • Figure 2: Distribution of papers over the years and number of rule- and model-based methods.
  • Figure :