Enhancing Hyperbole and Metaphor Detection with Their Bidirectional Dynamic Interaction and Emotion Knowledge

2025-06-18

AI Summary

Rhetorical figure identification—specifically hyperbole and metaphor—is highly challenging due to semantic ambiguity, expressive diversity, and implicit affective underpinnings. Existing approaches predominantly rely on shallow linguistic features, neglecting the interdependence between hyperbole and metaphor as well as their affective drivers. To address this, we propose EmoBi, an emotion-guided joint detection framework. EmoBi is the first to model bidirectional dynamic interaction between hyperbole and metaphor; it tightly couples fine-grained emotion analysis with source/target domain mapping to explicitly characterize how emotion modulates rhetorical perception; and it incorporates a detection verification module to enhance robustness. The framework comprises deep emotion mining, emotion-based domain mapping, a bidirectional interaction network, multi-dataset joint training, and a unified validation mechanism. Evaluated on four benchmarks—including TroFi and HYPO-L—EmoBi achieves new state-of-the-art performance: +28.1% F1 for hyperbole detection and +23.1% F1 for metaphor detection.

Technology Category

• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Search and Optimization: Metareasoning and Metaheuristics
• Cognitive Modeling & Cognitive Systems: Affective Computing

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

Abstract

Text-based hyperbole and metaphor detection are of great significance for natural language processing (NLP) tasks. However, due to their semantic obscurity and expressive diversity, it is rather challenging to identify them. Existing methods mostly focus on superficial text features, ignoring the associations of hyperbole and metaphor as well as the effect of implicit emotion on perceiving these rhetorical devices. To implement these hypotheses, we propose an emotion-guided hyperbole and metaphor detection framework based on bidirectional dynamic interaction (EmoBi). Firstly, the emotion analysis module deeply mines the emotion connotations behind hyperbole and metaphor. Next, the emotion-based domain mapping module identifies the target and source domains to gain a deeper understanding of the implicit meanings of hyperbole and metaphor. Finally, the bidirectional dynamic interaction module enables the mutual promotion between hyperbole and metaphor. Meanwhile, a verification mechanism is designed to ensure detection accuracy and reliability. Experiments show that EmoBi outperforms all baseline methods on four datasets. Specifically, compared to the current SoTA, the F1 score increased by 28.1% for hyperbole detection on the TroFi dataset and 23.1% for metaphor detection on the HYPO-L dataset. These results, underpinned by in-depth analyses, underscore the effectiveness and potential of our approach for advancing hyperbole and metaphor detection.

Problem

• Detect hyperbole and metaphor in text using emotion knowledge
• Address semantic obscurity and diversity in rhetorical devices
• Improve accuracy via bidirectional interaction and emotion analysis

Innovation

• Emotion-guided framework for hyperbole and metaphor detection
• Bidirectional dynamic interaction enhances mutual promotion
• Emotion-based domain mapping for implicit meaning understanding