Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis

Yusong Mu; Shumin Shi

doi:10.1007/978-3-031-57919-6_4

Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis

Yusong Mu^*, Shumin Shi

^*此作品的通讯作者

计算机学院

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

1 引用（Scopus）

摘要

Previous studies consider little on using dependency-type messages in the E2E-ABSA task. Studies using dependency-type messages just contact the dependency-type message and word embedding vectors, which may not fully fuse the context feature and information from the dependency type. This paper proposes a new model called Dependency-Type Weighted Graph Convolution Network (DTW-GCN) to compose dependency-type messages and word embedding. We use a type-weighted matrix to combine the dependency-type message, and DTW-GCN could fuse the dependency-type message and word embedding vectors. Experiments conducted on three benchmark datasets verify the effectiveness of our model.

源语言	英语
主期刊名	Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings
编辑	Zhongzhi Shi, Jim Torresen, Shengxiang Yang
出版商	Springer Science and Business Media Deutschland GmbH
页	46-57
页数	12
ISBN（印刷版）	9783031579189
DOI	https://doi.org/10.1007/978-3-031-57919-6_4
出版状态	已出版 - 2024
活动	13th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2024 - Shenzhen, 中国期限: 3 5月 2024 → 6 5月 2024

出版系列

姓名	IFIP Advances in Information and Communication Technology
卷	704 IFIPAICT
ISSN（印刷版）	1868-4238
ISSN（电子版）	1868-422X

会议

会议	13th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2024
国家/地区	中国
市	Shenzhen
时期	3/05/24 → 6/05/24

访问文件

10.1007/978-3-031-57919-6_4

其它文件与链接

链接到 Scopus 的出版物

引用此

Mu, Y., & Shi, S. (2024). Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis. 在 Z. Shi, J. Torresen, & S. Yang (编辑), Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings (页码 46-57). (IFIP Advances in Information and Communication Technology; 卷 704 IFIPAICT). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-57919-6_4

Mu, Yusong ; Shi, Shumin. / Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis. Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings. 编辑 / Zhongzhi Shi ; Jim Torresen ; Shengxiang Yang. Springer Science and Business Media Deutschland GmbH, 2024. 页码 46-57 (IFIP Advances in Information and Communication Technology).

@inproceedings{e49a683cbb4e4f5391da4ae470053192,

title = "Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis",

abstract = "Previous studies consider little on using dependency-type messages in the E2E-ABSA task. Studies using dependency-type messages just contact the dependency-type message and word embedding vectors, which may not fully fuse the context feature and information from the dependency type. This paper proposes a new model called Dependency-Type Weighted Graph Convolution Network (DTW-GCN) to compose dependency-type messages and word embedding. We use a type-weighted matrix to combine the dependency-type message, and DTW-GCN could fuse the dependency-type message and word embedding vectors. Experiments conducted on three benchmark datasets verify the effectiveness of our model.",

keywords = "Dependency syntactic knowledge, End-to-End Aspect-Based Sentiment Analysis, Graph Convolutional Network",

author = "Yusong Mu and Shumin Shi",

note = "Publisher Copyright: {\textcopyright} IFIP International Federation for Information Processing 2024.; 13th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2024 ; Conference date: 03-05-2024 Through 06-05-2024",

year = "2024",

doi = "10.1007/978-3-031-57919-6_4",

language = "English",

isbn = "9783031579189",

series = "IFIP Advances in Information and Communication Technology",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "46--57",

editor = "Zhongzhi Shi and Jim Torresen and Shengxiang Yang",

booktitle = "Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings",

address = "Germany",

}

Mu, Y & Shi, S 2024, Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis. 在 Z Shi, J Torresen & S Yang (编辑), Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings. IFIP Advances in Information and Communication Technology, 卷 704 IFIPAICT, Springer Science and Business Media Deutschland GmbH, 页码 46-57, 13th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2024, Shenzhen, 中国, 3/05/24. https://doi.org/10.1007/978-3-031-57919-6_4

Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis. / Mu, Yusong; Shi, Shumin.
Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings. 编辑 / Zhongzhi Shi; Jim Torresen; Shengxiang Yang. Springer Science and Business Media Deutschland GmbH, 2024. 页码 46-57 (IFIP Advances in Information and Communication Technology; 卷 704 IFIPAICT).

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

TY - GEN

T1 - Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis

AU - Mu, Yusong

AU - Shi, Shumin

N1 - Publisher Copyright: © IFIP International Federation for Information Processing 2024.

PY - 2024

Y1 - 2024

N2 - Previous studies consider little on using dependency-type messages in the E2E-ABSA task. Studies using dependency-type messages just contact the dependency-type message and word embedding vectors, which may not fully fuse the context feature and information from the dependency type. This paper proposes a new model called Dependency-Type Weighted Graph Convolution Network (DTW-GCN) to compose dependency-type messages and word embedding. We use a type-weighted matrix to combine the dependency-type message, and DTW-GCN could fuse the dependency-type message and word embedding vectors. Experiments conducted on three benchmark datasets verify the effectiveness of our model.

AB - Previous studies consider little on using dependency-type messages in the E2E-ABSA task. Studies using dependency-type messages just contact the dependency-type message and word embedding vectors, which may not fully fuse the context feature and information from the dependency type. This paper proposes a new model called Dependency-Type Weighted Graph Convolution Network (DTW-GCN) to compose dependency-type messages and word embedding. We use a type-weighted matrix to combine the dependency-type message, and DTW-GCN could fuse the dependency-type message and word embedding vectors. Experiments conducted on three benchmark datasets verify the effectiveness of our model.

KW - Dependency syntactic knowledge

KW - End-to-End Aspect-Based Sentiment Analysis

KW - Graph Convolutional Network

UR - http://www.scopus.com/inward/record.url?scp=85190857874&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-57919-6_4

DO - 10.1007/978-3-031-57919-6_4

M3 - Conference contribution

AN - SCOPUS:85190857874

SN - 9783031579189

T3 - IFIP Advances in Information and Communication Technology

SP - 46

EP - 57

BT - Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings

A2 - Shi, Zhongzhi

A2 - Torresen, Jim

A2 - Yang, Shengxiang

PB - Springer Science and Business Media Deutschland GmbH

T2 - 13th IFIP TC 12 International Conference on Intelligent Information Processing, IIP 2024

Y2 - 3 May 2024 through 6 May 2024

ER -

Mu Y, Shi S. Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis. 在 Shi Z, Torresen J, Yang S, 编辑, Intelligent Information Processing XII - 13th IFIP TC 12 International Conference, IIP 2024, Proceedings. Springer Science and Business Media Deutschland GmbH. 2024. 页码 46-57. (IFIP Advances in Information and Communication Technology). doi: 10.1007/978-3-031-57919-6_4

Dependency-Type Weighted Graph Convolutional Network on End-to-End Aspect-Based Sentiment Analysis

摘要

出版系列

会议

访问文件

其它文件与链接

指纹

引用此