Version 1
: Received: 23 April 2024 / Approved: 23 April 2024 / Online: 25 April 2024 (15:11:24 CEST)
How to cite:
Lv, S.; Dong, J.; Wang, C.; Wang, X.; Bao, Z. RB-GAT: A Text Classification Model Based on RoBERTa-BiGRU with Graph ATtention Network. Preprints2024, 2024041579. https://doi.org/10.20944/preprints202404.1579.v1
Lv, S.; Dong, J.; Wang, C.; Wang, X.; Bao, Z. RB-GAT: A Text Classification Model Based on RoBERTa-BiGRU with Graph ATtention Network. Preprints 2024, 2024041579. https://doi.org/10.20944/preprints202404.1579.v1
Lv, S.; Dong, J.; Wang, C.; Wang, X.; Bao, Z. RB-GAT: A Text Classification Model Based on RoBERTa-BiGRU with Graph ATtention Network. Preprints2024, 2024041579. https://doi.org/10.20944/preprints202404.1579.v1
APA Style
Lv, S., Dong, J., Wang, C., Wang, X., & Bao, Z. (2024). RB-GAT: A Text Classification Model Based on RoBERTa-BiGRU with Graph ATtention Network. Preprints. https://doi.org/10.20944/preprints202404.1579.v1
Chicago/Turabian Style
Lv, S., Xuanhong Wang and Zhiqiang Bao. 2024 "RB-GAT: A Text Classification Model Based on RoBERTa-BiGRU with Graph ATtention Network" Preprints. https://doi.org/10.20944/preprints202404.1579.v1
Abstract
With the development of deep learning, several Graph Neural Networks (GNN)-based approaches have been utilized for text classification. However, GNNs encounter challenges in capturing contextual text information within a document sequence. To address this, a novel text classification model RB-GAT is proposed by combining RoBERTa-BiGRU embedding and a multi-head Graph ATtention Network (GAT). First, the pre-trained RoBERTa model is exploited to learn word and text embeddings in different contexts. Second, the Bidirectional Gated Recurrent Unit (BiGRU) is employed to capture long-term dependencies and bidirectional sentence information from the text context. Next, the multi-head graph attention network is applied to analyze this information, which serves as a node feature for the document. Finally, the classification results are generated through a Softmax layer. Experimental results on three benchmark datasets demonstrate that our method can achieve an accuracy of 71.48%, 98.45%, and 80.32% on Ohsumed, R8, and MR, which is superior to the existing nine text classification approaches.
Keywords
word embedding; RoBERTa; BiGRU; text classification; multi-head GAT
Subject
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.