Details
| Original language | English |
|---|---|
| Number of pages | 19 |
| Journal | Journal of Intelligent Information Systems |
| Volume | 58 |
| Issue number | 1 |
| Early online date | 15 May 2021 |
| Publication status | Published - Feb 2022 |
Abstract
Fact checking is an essential challenge when combating fake news. Identifying documents that agree or disagree with a particular statement (claim) is a core task in this process. In this context, stance detection aims at identifying the position (stance) of a document towards a claim. Most approaches address this task through classification models that do not consider the highly imbalanced class distribution. Therefore, they are particularly ineffective in detecting the minority classes (for instance, ‘disagree’), even though such instances are crucial for tasks such as fact-checking by providing evidence for detecting false claims. In this paper, we exploit the hierarchical nature of stance classes which allows us to propose a modular pipeline of cascading binary classifiers, enabling performance tuning on a per step and class basis. We implement our approach through a combination of neural and traditional classification models that highlight the misclassification costs of minority classes. Evaluation results demonstrate state-of-the-art performance of our approach and its ability to significantly improve the classification performance of the important ‘disagree’ class.
Keywords
- Cascading classifiers, Fact-checking, Fake News, Stance detection
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Information Systems
- Computer Science(all)
- Hardware and Architecture
- Computer Science(all)
- Computer Networks and Communications
- Computer Science(all)
- Artificial Intelligence
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Journal of Intelligent Information Systems, Vol. 58, No. 1, 02.2022.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Exploiting stance hierarchies for cost-sensitive stance detection of Web documents
AU - Roy, Arjun
AU - Fafalios, Pavlos
AU - Ekbal, Asif
AU - Zhu, Xiaofei
AU - Dietze, Stefan
PY - 2022/2
Y1 - 2022/2
N2 - Fact checking is an essential challenge when combating fake news. Identifying documents that agree or disagree with a particular statement (claim) is a core task in this process. In this context, stance detection aims at identifying the position (stance) of a document towards a claim. Most approaches address this task through classification models that do not consider the highly imbalanced class distribution. Therefore, they are particularly ineffective in detecting the minority classes (for instance, ‘disagree’), even though such instances are crucial for tasks such as fact-checking by providing evidence for detecting false claims. In this paper, we exploit the hierarchical nature of stance classes which allows us to propose a modular pipeline of cascading binary classifiers, enabling performance tuning on a per step and class basis. We implement our approach through a combination of neural and traditional classification models that highlight the misclassification costs of minority classes. Evaluation results demonstrate state-of-the-art performance of our approach and its ability to significantly improve the classification performance of the important ‘disagree’ class.
AB - Fact checking is an essential challenge when combating fake news. Identifying documents that agree or disagree with a particular statement (claim) is a core task in this process. In this context, stance detection aims at identifying the position (stance) of a document towards a claim. Most approaches address this task through classification models that do not consider the highly imbalanced class distribution. Therefore, they are particularly ineffective in detecting the minority classes (for instance, ‘disagree’), even though such instances are crucial for tasks such as fact-checking by providing evidence for detecting false claims. In this paper, we exploit the hierarchical nature of stance classes which allows us to propose a modular pipeline of cascading binary classifiers, enabling performance tuning on a per step and class basis. We implement our approach through a combination of neural and traditional classification models that highlight the misclassification costs of minority classes. Evaluation results demonstrate state-of-the-art performance of our approach and its ability to significantly improve the classification performance of the important ‘disagree’ class.
KW - Cascading classifiers
KW - Fact-checking
KW - Fake News
KW - Stance detection
UR - http://www.scopus.com/inward/record.url?scp=85106289190&partnerID=8YFLogxK
U2 - 10.1007/s10844-021-00642-z
DO - 10.1007/s10844-021-00642-z
M3 - Article
AN - SCOPUS:85106289190
VL - 58
JO - Journal of Intelligent Information Systems
JF - Journal of Intelligent Information Systems
SN - 0925-9902
IS - 1
ER -