Stylistic analysis of translated languages: A perturbation-based XAI deep learning framework
PLOS One · 07.07.2026 · 21 (7) · e0352889
DOI: 10.1371/journal.pone.0352889 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 07.07.2026. Verfasst von Dan Feng Huang und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Foreign Languages, Guangdong Polytechnic Normal University, Guangdong, China. Hauptthemen laut MEDLINE: Deep Learning, Natural Language Processing, Translating, Language. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Nanyang Technological University, Hong Kong University of Science and Technology.
Abstract (Originalsprache)
Text classification using traditional machine learning techniques has been used in natural language processing (NLP) tasks to distinguish translated from non-translated languages, with high accuracy scores indicating the distinctive style of translated languages. While deep learning (DL) has demonstrated impressive performance in terms of representation learning and capturing nuanced patterns in natural language data, DL models act as black boxes, making their results difficult to interpret. This study addresses this issue by demonstrating an explainable AI (XAI) DL framework in a case study of United Nations (UN) meetings. The framework consists of three stages: i) train a variational autoencoder (VAE) combined with BERT embeddings converted from translated and non-translated texts; ii) utilize the majority vote from three classifiers selected from a stacked ensemble to classify the VAE’s latent representations; iii) implement a perturbation-based XAI method to interpret the DL model’s decisions. The results indicate that the VAE-based model effectively distinguishes the two text types, with accuracy scores above 0.8. The XAI analysis reveals that interpreting the VAE-based model’s decision uncovers stylistic differences between the two text types beyond superficial lexical and syntactic features. This proof-of-concept study demonstrates the potential of the XAI DL framework in other NLP studies that aim to analyze style.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Foreign Languages, Guangdong Polytechnic Normal University, Guangdong, China.
- Divison of Humanities, The Hong Kong University of Science and Technology, Hong Kong SAR, China.
ORCID-Profile: Dan Feng Huang
Förderung
- Nanyang Technological University
- Hong Kong University of Science and Technology
Zitieren
APA
Huang, D. F., & Tay, D. (2026). Stylistic analysis of translated languages: A perturbation-based XAI deep learning framework. PLOS One, 21(7), e0352889. https://doi.org/10.1371/journal.pone.0352889
Vancouver
Huang DF, Tay D. Stylistic analysis of translated languages: A perturbation-based XAI deep learning framework. PLOS One. 2026;21(7):e0352889. doi:10.1371/journal.pone.0352889
BibTeX
@article{Huang2026,
author = {Huang, Dan Feng and Tay, Dennis},
title = {Stylistic analysis of translated languages: A perturbation-based XAI deep learning framework},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0352889},
doi = {10.1371/journal.pone.0352889}
}
EUVIMED