Core Concepts
Emotion Analysis in NLP is evolving but faces challenges in defining tasks, frameworks, and cultural considerations.
Abstract
Introduction
Emotions play a crucial role in human cognition and relationships.
Interest in emotion analysis (EA) in NLP has surged since 2018.
A review of over 150 ACL papers on EA was conducted.
Related Work
Various surveys provide an overview of the state of the field.
Recent surveys focus on datasets, models, and applications.
Survey of Emotion Analysis in NLP
Selection of studies based on EA keywords.
Analysis of papers published in the last decade.
Overview of datasets used in EA studies.
Shortcomings of EA in NLP
Lack of diversity in available datasets.
Inconsistency in terminology.
Lack of interdisciplinarity.
Demographics and cultural implications.
Discussion
EA tasks, emotion frameworks, cultural considerations, and applications.
Proposed future directions for diversity, tailored emotion categories, nomenclature, and interdisciplinarity.
Roadmap for Future Directions
Collecting demographic data for diversity.
Tailoring emotion categories to specific tasks.
Standardizing EA nomenclature.
Encouraging interdisciplinary engagement.
Conclusion
Growth of EA in NLP with increasing research on human emotions.
Review of ACL papers on EA and identified gaps and future directions.
Stats
Emotion Analysis (EA) in NLP has surged since 2018.
Over 150 ACL papers on EA were reviewed.
Various emotion models and frameworks are used in EA studies.
Quotes
"Emotions perfume our every experience and interaction."
"EA has attracted considerable research interest in recent decades."