CASIA OpenIR  > 多模态人工智能系统全国重点实验室  > 自然语言处理
CNN-Based Broad Learning for Cross-Domain Emotion Classification
Zeng, Rong1; Liu, Hongzhan1; Peng, Sancheng2; Cao, Lihong2; Yang, Aimin3; Zong, Chengqing4; Zhou, Guodong5
Source PublicationTSINGHUA SCIENCE AND TECHNOLOGY
ISSN1007-0214
2023-04-01
Volume28Issue:2Pages:360-369
Corresponding AuthorLiu, Hongzhan(lhzscnu@163.com) ; Peng, Sancheng(psc346@aliyun.com)
AbstractCross-domain emotion classification aims to leverage useful information in a source domain to help predict emotion polarity in a target domain in a unsupervised or semi-supervised manner. Due to the domain discrepancy, an emotion classifier trained on source domain may not work well on target domain. Many researchers have focused on traditional cross-domain sentiment classification, which is coarse-grained emotion classification. However, the problem of emotion classification for cross-domain is rarely involved. In this paper, we propose a method, called convolutional neural network (CNN) based broad learning, for cross-domain emotion classification by combining the strength of CNN and broad learning. We first utilized CNN to extract domain-invariant and domain-specific features simultaneously, so as to train two more efficient classifiers by employing broad learning. Then, to take advantage of these two classifiers, we designed a co-training model to boost together for them. Finally, we conducted comparative experiments on four datasets for verifying the effectiveness of our proposed method. The experimental results show that the proposed method can improve the performance of emotion classification more effectively than those baseline methods.
KeywordMeasurement Deep learning Adaptation models Feature extraction Convolutional neural networks Data mining Task analysis cross-domain emotion classification CNN broad learning classifier co-training
DOI10.26599/TST.2022.9010007
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Information Systems ; Computer Science, Software Engineering ; Engineering, Electrical & Electronic
WOS IDWOS:000862392800014
PublisherTSINGHUA UNIV PRESS
Citation statistics
Cited Times:8[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/50401
Collection多模态人工智能系统全国重点实验室_自然语言处理
Corresponding AuthorLiu, Hongzhan; Peng, Sancheng
Affiliation1.South China Normal Univ, Guangdong Prov Key Lab Nanophoton Funct Mat & Dev, Guangzhou 511400, Peoples R China
2.Guangdong Univ Foreign Studies, Lab Language Engn & Comp, Guangzhou 510006, Peoples R China
3.Lingnan Normal Univ, Sch Comp Sci & Intelligence Educ, Guangzhou 510006, Peoples R China
4.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
5.Soochow Univ, Sch Comp Sci & Technol, Suzhou 215031, Peoples R China
Recommended Citation
GB/T 7714
Zeng, Rong,Liu, Hongzhan,Peng, Sancheng,et al. CNN-Based Broad Learning for Cross-Domain Emotion Classification[J]. TSINGHUA SCIENCE AND TECHNOLOGY,2023,28(2):360-369.
APA Zeng, Rong.,Liu, Hongzhan.,Peng, Sancheng.,Cao, Lihong.,Yang, Aimin.,...&Zhou, Guodong.(2023).CNN-Based Broad Learning for Cross-Domain Emotion Classification.TSINGHUA SCIENCE AND TECHNOLOGY,28(2),360-369.
MLA Zeng, Rong,et al."CNN-Based Broad Learning for Cross-Domain Emotion Classification".TSINGHUA SCIENCE AND TECHNOLOGY 28.2(2023):360-369.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zeng, Rong]'s Articles
[Liu, Hongzhan]'s Articles
[Peng, Sancheng]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zeng, Rong]'s Articles
[Liu, Hongzhan]'s Articles
[Peng, Sancheng]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zeng, Rong]'s Articles
[Liu, Hongzhan]'s Articles
[Peng, Sancheng]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.