Skip to content

Read-only version: For security reasons, the Research Information System (FIS) has been temporarily replaced by this read-only version. All public content and full texts remain available; login, editing, submission and statistics are not available. Questions: fis@uni-bamberg.de

Home › Publications › Topic Bias in Emotion Classification

Publication

Topic Bias in Emotion Classification

Wegge, Maximilian; Klinger, Roman (2024): Topic Bias in Emotion Classification, in: Rob van der Goot, JinYeong Bak, Max Müller-Eberstein, u. a. (Hrsg.), Proceedings of the Ninth Workshop on Noisy and User-generated Text (W-NUT 2024), San Ġiljan, Malta: Association for Computational Linguistics, S. 89–103.

Details

Faculty/Chair
Faculty/Professorship
Author
Wegge, Maximilian
Title of the compilation
Proceedings of the Ninth Workshop on Noisy and User-generated Text (W-NUT 2024)
Editors
Goot, Rob van der
Bak, JinYeong
Müller-Eberstein, Max
Xu, Wei
Ritter, Alan
Baldwin, Tim
Conference
Ninth Workshop on Noisy and User-generated Text (W-NUT 2024), San Ġiljan, Malta
Publisher Information
Association for Computational Linguistics : San Ġiljan, Malta
Year of publication
2024
Language
English
URL
Abstract

Emotion corpora are typically sampled based on keyword/hashtag search or by asking study participants to generate textual instances. In any case, these corpora are not uniform samples representing the entirety of a domain. We hypothesize that this practice of data acquision leads to unrealistic correlations between overrepresented topics in these corpora that harm the generalizability of models. Such topic bias could lead to wrong predictions for instances like “I organized the service for my aunt’s funeral.” when funeral events are overpresented for instances labeled with sadness, despite the emotion of pride being more appropriate here. In this paper, we study this topic bias both from the data and the modeling perspective. We first label a set of emotion corpora automatically via topic modeling and show that emotions in fact correlate with specific topics. Further, we see that emotion classifiers are confounded by such topics. Finally, we show that the established debiasing method of adversarial correction via gradient reversal mitigates the issue. Our work points out issues with existing emotion corpora and that more representative resources are required for fair evaluation of models predicting affective concepts from text.

GND Keywords
Computerlinguistik
Gefühl
Klassifikation
Keywords
emotion classification
DDC Classification
004 Computer science
RVK Classification
ST 306
Type
Conferenceobject
URI
Activation date
2024-03-07
Project(s)

Permalink: https://fis.uni-bamberg.de/handle/uniba/93869 · Data as JSON · Cite: BibTeX · RIS

Data as of: 2026-11-15

Restricted read-only access: login, editing and submission are temporarily disabled.