A Domain-adaptive Pre-training Approach for Language Bias Detection in News

被引:6
|
作者
Krieger, Jan-David [1 ]
Spinde, Timo [2 ]
Ruas, Terry [2 ]
Kulshrestha, Juhi [1 ]
Gipp, Bela [3 ]
机构
[1] Univ Konstanz, Constance, Germany
[2] Univ Wuppertal, Wuppertal, Germany
[3] Univ Gottingen, Gottingen, Germany
关键词
Media bias; news slant; neural classification; text analysis; domain adaptive;
D O I
10.1145/3529372.3530932
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Media bias is a multi-faceted construct influencing individual behavior and collective decision-making. Slanted news reporting is the result of one-sided and polarized writing which can occur in various forms. In this work, we focus on an important form of media bias, i.e. bias by word choice. Detecting biased word choices is a challenging task due to its linguistic complexity and the lack of representative gold-standard corpora. We present DA-RoBERTa, a new state-of-the-art transformer-based model adapted to the media bias domain which identifies sentence-level bias with an F1 score of 0.814. In addition, we also train, DA-BERT and DA-BART, two more transformer models adapted to the bias domain. Our proposed domain-adapted models outperform prior bias detection approaches on the same data.
引用
收藏
页数:7
相关论文
共 50 条
  • [1] Domain-adaptive pre-training on a BERT model for the automatic detection of misogynistic tweets in Spanish
    Rodriguez, Dalia A.
    Diaz-Escobar, Julia
    Diaz-Ramirez, Arnoldo
    Trujillo, Leonardo
    [J]. SOCIAL NETWORK ANALYSIS AND MINING, 2023, 13 (01)
  • [2] Domain-adaptive pre-training on a BERT model for the automatic detection of misogynistic tweets in Spanish
    Dalia A. Rodríguez
    Julia Diaz-Escobar
    Arnoldo Díaz-Ramírez
    Leonardo Trujillo
    [J]. Social Network Analysis and Mining, 13
  • [3] K-DLM: A Domain-Adaptive Language Model Pre-Training Framework with Knowledge Graph
    Zou, Jiaxin
    Xie, Zuotong
    Chen, Junhua
    Hou, Jiawei
    Yan, Qiang
    Zheng, Hai-Tao
    [J]. ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT IV, 2023, 14257 : 447 - 459
  • [4] Improving Linguistic Bias Detection in Wikipedia using Cross-Domain Adaptive Pre-Training
    Madanagopal, Karthic
    Caverlee, James
    [J]. COMPANION PROCEEDINGS OF THE WEB CONFERENCE 2022, WWW 2022 COMPANION, 2022, : 1301 - 1309
  • [5] Diagnosing crop diseases based on domain-adaptive pre-training BERT of electronic medical records
    Ding, Junqi
    Li, Bo
    Xu, Chang
    Qiao, Yan
    Zhang, Lingxian
    [J]. APPLIED INTELLIGENCE, 2023, 53 (12) : 15979 - 15992
  • [6] Empowering crisis information extraction through actionability event schemata and domain-adaptive pre-training
    Zhang, Yuhao
    Lo, Siaw Ling
    Myint, Phyo Yi Win
    [J]. Information and Management, 2025, 62 (01):
  • [7] Diagnosing crop diseases based on domain-adaptive pre-training BERT of electronic medical records
    Junqi Ding
    Bo Li
    Chang Xu
    Yan Qiao
    Lingxian Zhang
    [J]. Applied Intelligence, 2023, 53 : 15979 - 15992
  • [8] Enhancing abusive language detection: A domain-adapted approach leveraging BERT pre-training tasks
    Jarquín-Vásquez, Horacio
    Escalante, Hugo Jair
    Montes-y-Gómez, Manuel
    [J]. Pattern Recognition Letters, 2024, 186 : 361 - 368
  • [9] Domain-Adaptive Fall Detection Using Deep Adversarial Training
    Liu, Kai-Chun
    Chan, Michael
    Kuo, Heng-Cheng
    Hsieh, Chia-Yeh
    Huang, Hsiang-Yun
    Chan, Chia-Tai
    Tsao, Yu
    [J]. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2021, 29 : 1243 - 1251
  • [10] Efficient Medical Images Text Detection with Vision-Language Pre-training Approach
    Li, Tianyang
    Bai, Jinxu
    Wang, Qingzhu
    Xu, Hanwen
    [J]. ASIAN CONFERENCE ON MACHINE LEARNING, VOL 222, 2023, 222