Saltar al contenido principal

ETHNOS_APP

Inicio • Búsqueda • Revistas • Lista 0

Unlocking Bias Detection

Leveraging Transformer-Based Models for Content Analysis

Datos Bibliográficos

ID22107366
AutoresShaina Raza (0000-0003-1061-5845, Vector Institute), Oluwanifemi Bamgbose (Vector Institute), Veronica Chatrath (0000-0003-4790-3963, Vector Institute), Shardule Ghuge (0009-0001-0728-0733, Vector Institute), Yan Sidyakin (Systems, Applications & Products in Data Processing (Canada)), Abdullah Yahya Mohammed Muaad (0000-0001-8304-9261, University of Mysore)
Año2024
Volumen11
Número5
Páginas6422-6434
Fecha de publicación2024-10-01
Peer ReviewedSí
Open AccessSí
TipoARTICLE
RevistaIEEE Transactions on Computational Social Systems (JOURNAL)
Identificadores de la revistaISSN: 2329-924X • E-ISSN: 2373-7476
EditorialInstitute of Electrical and Electronics Engineers (IEEE) (PUBLISHER)
DOI10.1109/tcss.2024.3392469
OpenAlexW4398788568
IdiomaEN
Referencias citadas35

Bias detection in text is crucial for combating the spread of negative stereotypes, misinformation, and biased decision-making. Traditional language models frequently face challenges in generalizing beyond their training data and are typically designed for a single task, often focusing on bias detection at the sentence level. To address this, we present the contextualized bi-directional dual transformer (CBDT) classifier. This model combines two complementary transformer networks: the context transformer and the entity transformer, with a focus on improving bias detection capabilities. We have prepared a dataset specifically for training these models to identify and locate biases in texts. Our evaluations across various datasets demonstrate CBDT effectiveness in distinguishing biased narratives from neutral ones and identifying specific biased terms. This work paves the way for applying the CBDT model in various linguistic and cultural contexts, enhancing its utility in bias detection efforts. We also make the annotated dataset available for research purposes

Machine learning · Natural language processing · Sentence · Transformer · Computer Science · Engineering · Natural Language Processing Techniques · Text Readability and Simplification · Topic Modeling · Artificial Intelligence

  • Persistent Anti-Muslim Bias in Large Language Models

    Open Access•Abubakar Abid, Maheen Farooqi et al.•Proceedings of the 2021 AAAI/ACM…•2021

  • Semantics derived automatically from language corpora contain human-like biases

    Open Access•Aysun Caliskan, Joanna J Bryson et al.•Science•2017

  • A Survey on Evaluation of Large Language Models

    Open Access•Yupeng Chang, Xu Wang et al.•ACM Transactions on Intelligent…•2024

  • Fast unfolding of communities in large networks

    Open Access•Vincent D Blondel, Jean-Loup Guillaume et al.•Journal of Statistical Mechanics:…•2008

  • Bias and Fairness in Large Language Models

    Open Access•Isabel O Gallegos, Ryan A Rossi et al.•Computational Linguistics•2024

  • A survey of named entity recognition and classification

    David R Nadeau, David Nadeau et al.•Lingvisticae investigationes.…•2007

Velocidad de citaciónhistorical
Altamente citadoNo
Ethnos_APP • Proyecto Open Source • Licencia MIT • Frontend v2.0.0 • Privacidad y Cookies • Documentación de la API: api.ethnos.app/docs • Código de la API: GitHub • DOI: 10.5281/zenodo.17049435 • Código del Frontend: GitHub • DOI: 10.5281/zenodo.17050053 • cruz.rio.br • Expectantes Misericordiae