Cargando…

A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism

Finding the news of same case from the large numbers of case-involved news is an important basis for public opinion analysis. Existing text clustering methods usually based on topic models which only use topic and case infomation as the global features of documents, so distinguishing between differe...

Descripción completa

Detalles Bibliográficos
Autores principales: Mao, Cunli, Liang, Haoyuan, Yu, Zhengtao, Huang, Yuxin, Guo, Junjun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8620995/
https://www.ncbi.nlm.nih.gov/pubmed/34833580
http://dx.doi.org/10.3390/s21227501
_version_ 1784605350898434048
author Mao, Cunli
Liang, Haoyuan
Yu, Zhengtao
Huang, Yuxin
Guo, Junjun
author_facet Mao, Cunli
Liang, Haoyuan
Yu, Zhengtao
Huang, Yuxin
Guo, Junjun
author_sort Mao, Cunli
collection PubMed
description Finding the news of same case from the large numbers of case-involved news is an important basis for public opinion analysis. Existing text clustering methods usually based on topic models which only use topic and case infomation as the global features of documents, so distinguishing between different cases with similar types remains a challenge. The contents of documents contain rich local features. Taking into account the internal features of news, the information of cases and the contributions provided by different topics, we propose a clustering method of case-involved news, which combines topic network and multi-head attention mechanism. Using case information and topic information to construct a topic network, then extracting the global features by graph convolution network, thus realizing the combination of case information and topic information. At the same time, the local features are extracted by multi-head attention mechanism. Finally, the fusion of global features and local features is realized by variational auto-encoder, and the learned latent representations are used for clustering. The experiments show that the proposed method significantly outperforms the state-of-the-art unsupervised clustering methods.
format Online
Article
Text
id pubmed-8620995
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-86209952021-11-27 A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism Mao, Cunli Liang, Haoyuan Yu, Zhengtao Huang, Yuxin Guo, Junjun Sensors (Basel) Article Finding the news of same case from the large numbers of case-involved news is an important basis for public opinion analysis. Existing text clustering methods usually based on topic models which only use topic and case infomation as the global features of documents, so distinguishing between different cases with similar types remains a challenge. The contents of documents contain rich local features. Taking into account the internal features of news, the information of cases and the contributions provided by different topics, we propose a clustering method of case-involved news, which combines topic network and multi-head attention mechanism. Using case information and topic information to construct a topic network, then extracting the global features by graph convolution network, thus realizing the combination of case information and topic information. At the same time, the local features are extracted by multi-head attention mechanism. Finally, the fusion of global features and local features is realized by variational auto-encoder, and the learned latent representations are used for clustering. The experiments show that the proposed method significantly outperforms the state-of-the-art unsupervised clustering methods. MDPI 2021-11-11 /pmc/articles/PMC8620995/ /pubmed/34833580 http://dx.doi.org/10.3390/s21227501 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Mao, Cunli
Liang, Haoyuan
Yu, Zhengtao
Huang, Yuxin
Guo, Junjun
A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title_full A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title_fullStr A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title_full_unstemmed A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title_short A Clustering Method of Case-Involved News by Combining Topic Network and Multi-Head Attention Mechanism
title_sort clustering method of case-involved news by combining topic network and multi-head attention mechanism
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8620995/
https://www.ncbi.nlm.nih.gov/pubmed/34833580
http://dx.doi.org/10.3390/s21227501
work_keys_str_mv AT maocunli aclusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT lianghaoyuan aclusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT yuzhengtao aclusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT huangyuxin aclusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT guojunjun aclusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT maocunli clusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT lianghaoyuan clusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT yuzhengtao clusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT huangyuxin clusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism
AT guojunjun clusteringmethodofcaseinvolvednewsbycombiningtopicnetworkandmultiheadattentionmechanism