Bert-Enhanced Text Graph Neural Network for Classification

Text classification is a fundamental research direction, aims to assign tags to text units. Recently, graph neural networks (GNN) have exhibited some excellent properties in textual information processing. Furthermore, the pre-trained language model also realized promising effects in many tasks. How...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yiping Yang, Xiaohui Cui
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	text classification Bert graph neural networks Science Q Astrophysics QB460-466 Physics QC1-999
Acceso en línea:	https://doaj.org/article/431c920053864371a6b3ab0e37f1febe
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:431c920053864371a6b3ab0e37f1febe
record_format	dspace
spelling	oai:doaj.org-article:431c920053864371a6b3ab0e37f1febe2021-11-25T17:30:41ZBert-Enhanced Text Graph Neural Network for Classification10.3390/e231115361099-4300https://doaj.org/article/431c920053864371a6b3ab0e37f1febe2021-11-01T00:00:00Zhttps://www.mdpi.com/1099-4300/23/11/1536https://doaj.org/toc/1099-4300Text classification is a fundamental research direction, aims to assign tags to text units. Recently, graph neural networks (GNN) have exhibited some excellent properties in textual information processing. Furthermore, the pre-trained language model also realized promising effects in many tasks. However, many text processing methods cannot model a single text unit’s structure or ignore the semantic features. To solve these problems and comprehensively utilize the text’s structure information and semantic information, we propose a Bert-Enhanced text Graph Neural Network model (BEGNN). For each text, we construct a text graph separately according to the co-occurrence relationship of words and use GNN to extract text features. Moreover, we employ Bert to extract semantic features. The former part can take into account the structural information, and the latter can focus on modeling the semantic information. Finally, we interact and aggregate these two features of different granularity to get a more effective representation. Experiments on standard datasets demonstrate the effectiveness of BEGNN.Yiping YangXiaohui CuiMDPI AGarticletext classificationBertgraph neural networksScienceQAstrophysicsQB460-466PhysicsQC1-999ENEntropy, Vol 23, Iss 1536, p 1536 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	text classification Bert graph neural networks Science Q Astrophysics QB460-466 Physics QC1-999
spellingShingle	text classification Bert graph neural networks Science Q Astrophysics QB460-466 Physics QC1-999 Yiping Yang Xiaohui Cui Bert-Enhanced Text Graph Neural Network for Classification
description	Text classification is a fundamental research direction, aims to assign tags to text units. Recently, graph neural networks (GNN) have exhibited some excellent properties in textual information processing. Furthermore, the pre-trained language model also realized promising effects in many tasks. However, many text processing methods cannot model a single text unit’s structure or ignore the semantic features. To solve these problems and comprehensively utilize the text’s structure information and semantic information, we propose a Bert-Enhanced text Graph Neural Network model (BEGNN). For each text, we construct a text graph separately according to the co-occurrence relationship of words and use GNN to extract text features. Moreover, we employ Bert to extract semantic features. The former part can take into account the structural information, and the latter can focus on modeling the semantic information. Finally, we interact and aggregate these two features of different granularity to get a more effective representation. Experiments on standard datasets demonstrate the effectiveness of BEGNN.
format	article
author	Yiping Yang Xiaohui Cui
author_facet	Yiping Yang Xiaohui Cui
author_sort	Yiping Yang
title	Bert-Enhanced Text Graph Neural Network for Classification
title_short	Bert-Enhanced Text Graph Neural Network for Classification
title_full	Bert-Enhanced Text Graph Neural Network for Classification
title_fullStr	Bert-Enhanced Text Graph Neural Network for Classification
title_full_unstemmed	Bert-Enhanced Text Graph Neural Network for Classification
title_sort	bert-enhanced text graph neural network for classification
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/431c920053864371a6b3ab0e37f1febe
work_keys_str_mv	AT yipingyang bertenhancedtextgraphneuralnetworkforclassification AT xiaohuicui bertenhancedtextgraphneuralnetworkforclassification
_version_	1718412280787369984

Bert-Enhanced Text Graph Neural Network for Classification

Ejemplares similares