Evaluating Deception Detection Model Robustness To Linguistic Variation

Maria Glenski, Ellyn Ayton, Robin Cosbey, Dustin Arendt, Svitlana Volkova


Abstract
With the increasing use of machine-learning driven algorithmic judgements, it is critical to develop models that are robust to evolving or manipulated inputs. We propose an extensive analysis of model robustness against linguistic variation in the setting of deceptive news detection, an important task in the context of misinformation spread online. We consider two prediction tasks and compare three state-of-the-art embeddings to highlight consistent trends in model performance, high confidence misclassifications, and high impact failures. By measuring the effectiveness of adversarial defense strategies and evaluating model susceptibility to adversarial attacks using character- and word-perturbed text, we find that character or mixed ensemble models are the most effective defenses and that character perturbation-based attack tactics are more successful.
Anthology ID:
2021.socialnlp-1.6
Volume:
Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media
Month:
June
Year:
2021
Address:
Online
Venues:
NAACL | SocialNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
70–80
Language:
URL:
https://aclanthology.org/2021.socialnlp-1.6
DOI:
10.18653/v1/2021.socialnlp-1.6
Bibkey:
Cite (ACL):
Maria Glenski, Ellyn Ayton, Robin Cosbey, Dustin Arendt, and Svitlana Volkova. 2021. Evaluating Deception Detection Model Robustness To Linguistic Variation. In Proceedings of the Ninth International Workshop on Natural Language Processing for Social Media, pages 70–80, Online. Association for Computational Linguistics.
Cite (Informal):
Evaluating Deception Detection Model Robustness To Linguistic Variation (Glenski et al., SocialNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.socialnlp-1.6.pdf