Details
| Original language | English |
|---|---|
| Pages (from-to) | 141-150 |
| Number of pages | 10 |
| Journal | NEUROCOMPUTING |
| Volume | 464 |
| Early online date | 25 Aug 2021 |
| Publication status | Published - 13 Nov 2021 |
| Externally published | Yes |
Abstract
Microblog text is usually very short, thereby challenging existing sentiment classification methods by providing models with little context. Recently, historical user information has been widely used in many real-world applications, such as recommender systems. However, few research works consider user historical states in the loop of microblog sentiment analysis. In this work, we propose to involve historical user information for microblog sentiment analysis to alleviate the context sparsity problem. In particular, we propose a novel neural microblog sentiment classification method which learns informative representations of microblog posts by exploiting both a user's contextual information and his/her historical state information. The proposed method consists of four components, i.e., a micropost encoder, a user historical sentiment encoder, a User Historical Semantic Encoder, and a micropost sentiment classification component. Extensive experiments are conducted on real-world data collected from Weibo, and experimental results show that the proposed approach achieves superior performance as compared to state-of-the-art baselines.
Keywords
- Microblog analysis, Sentiment classification, User historical preference
ASJC Scopus subject areas
- Computer Science(all)
- Computer Science Applications
- Neuroscience(all)
- Cognitive Neuroscience
- Computer Science(all)
- Artificial Intelligence
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: NEUROCOMPUTING, Vol. 464, 13.11.2021, p. 141-150.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Exploring user historical semantic and sentiment preference for microblog sentiment classification
AU - Zhu, Xiaofei
AU - Wu, Jie
AU - Zhu, Ling
AU - Guo, Jiafeng
AU - Yu, Ran
AU - Boland, Katarina
AU - Dietze, Stefan
N1 - Funding Information: This work was supported by the National Natural Science Foundation of China (No. 61722211 ); the Federal Ministry of Education and Research (No. 01LE1806A ); the Beijing Academy of Artificial Intelligence (No. BAAI2019ZD0306 ); the Technology Innovation and Application Development of Chongqing (No. cstc2020jscx-dxwtBX0014 ).
PY - 2021/11/13
Y1 - 2021/11/13
N2 - Microblog text is usually very short, thereby challenging existing sentiment classification methods by providing models with little context. Recently, historical user information has been widely used in many real-world applications, such as recommender systems. However, few research works consider user historical states in the loop of microblog sentiment analysis. In this work, we propose to involve historical user information for microblog sentiment analysis to alleviate the context sparsity problem. In particular, we propose a novel neural microblog sentiment classification method which learns informative representations of microblog posts by exploiting both a user's contextual information and his/her historical state information. The proposed method consists of four components, i.e., a micropost encoder, a user historical sentiment encoder, a User Historical Semantic Encoder, and a micropost sentiment classification component. Extensive experiments are conducted on real-world data collected from Weibo, and experimental results show that the proposed approach achieves superior performance as compared to state-of-the-art baselines.
AB - Microblog text is usually very short, thereby challenging existing sentiment classification methods by providing models with little context. Recently, historical user information has been widely used in many real-world applications, such as recommender systems. However, few research works consider user historical states in the loop of microblog sentiment analysis. In this work, we propose to involve historical user information for microblog sentiment analysis to alleviate the context sparsity problem. In particular, we propose a novel neural microblog sentiment classification method which learns informative representations of microblog posts by exploiting both a user's contextual information and his/her historical state information. The proposed method consists of four components, i.e., a micropost encoder, a user historical sentiment encoder, a User Historical Semantic Encoder, and a micropost sentiment classification component. Extensive experiments are conducted on real-world data collected from Weibo, and experimental results show that the proposed approach achieves superior performance as compared to state-of-the-art baselines.
KW - Microblog analysis
KW - Sentiment classification
KW - User historical preference
UR - http://www.scopus.com/inward/record.url?scp=85114128834&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2021.08.089
DO - 10.1016/j.neucom.2021.08.089
M3 - Article
AN - SCOPUS:85114128834
VL - 464
SP - 141
EP - 150
JO - NEUROCOMPUTING
JF - NEUROCOMPUTING
SN - 0925-2312
ER -