Detection of AI-generated Texts: A Bi-LSTM and Attention-Based Approach

Abstract

This paper presents a novel algorithm that leverages cutting-edge machine-learning techniques to accurately and efficiently detect AI-generated texts. Rapid advancements in natural language processing models have led to the generation of text closely resembling human language, making it increasingly difficult to differentiate between human and AI-generated content. However, misuse of such texts presents a serious and imminent threat to the quality of academic publishing. This underscores the urgent need for robust detection mechanisms to ensure information quality, maintain trust, and preserve the integrity of research publications. Our proposed model outperformed existing algorithms for accuracy with less computational complexity. The proposed model is a feature-based hybrid deep learning network that leverages part-of-speech tagging and integrates Bidirectional Long Short-Term Memory (BiLSTM) networks with Attention modules. The initial module extracts local contextual features using convolutional layers, followed by BiLSTM layers that capture long-term dependencies from past and future sequences. An attention mechanism highlights critical sequence components, enhancing the model’s focus on relevant data. The outputs from the attention and initial modules are concatenated through a residual connection, ensuring comprehensive feature representation. This combination is then fed into dense layers for final classification, effectively balancing feature richness and computational efficiency. The proposed model was evaluated on two benchmark datasets, achieving 85.00% and 88.00% accuracy, respectively.

Publication DOI: https://doi.org/10.1109/ACCESS.2025.3562750
Divisions: College of Business and Social Sciences > School of Social Sciences & Humanities > Centre for Language Research at Aston (CLaRA)
College of Business and Social Sciences > Aston Business School > Cyber Security Innovation (CSI) Research Centre
College of Business and Social Sciences > School of Social Sciences & Humanities > English Languages and Applied Linguistics
?? 53981500Jl ??
College of Business and Social Sciences > Aston Institute for Forensic Linguistics
Funding Information: This work was supported in part by Seed-Corn funding received from the Aston Institute for Forensic Linguistics.
Additional Information: Copyright © 2025 The Authors. This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
Uncontrolled Keywords: Accuracy,Detectors,Computational modeling,Transformers,Text detection,Feature extraction,Context modeling,Attention mechanisms,Fake news,Deep learning,AI-generated text detection,authorship analysis,authorship verification,machine-generated text detection
Publication ISSN: 2169-3536
Last Modified: 07 May 2025 16:01
Date Deposited: 24 Apr 2025 12:44
Full Text Link:
Related URLs: https://ieeexpl ... ument/10971184/ (Publisher URL)
http://www.scop ... tnerID=8YFLogxK (Scopus URL)
PURE Output Type: Article
Published Date: 2025-04-21
Published Online Date: 2025-04-21
Accepted Date: 2025-04-10
Authors: Blake, John
Miah, Abu Saleh Musa
Kredens, Krzysztof (ORCID Profile 0000-0001-7038-9478)
Shin, Jungpil

Download

[img]

Version: Accepted Version

Access Restriction: Restricted to Repository staff only

License: Creative Commons Attribution


[img]

Version: Published Version

License: Creative Commons Attribution


Export / Share Citation


Statistics

Additional statistics for this record