Few-shot Aspect Category Sentiment Analysis via Meta-learning

Bin Liang; Xiang Li; Lin Gui; Yonghao Fu; Yulan He; Min Yang; Ruifeng Xu

doi:10.1145/3529954

Few-shot Aspect Category Sentiment Analysis via Meta-learning

Bin Liang, Xiang Li, Lin Gui, Yonghao Fu, Yulan He, Min Yang, Ruifeng Xu^*

^*Corresponding author for this work

Informatics

Research output: Contribution to journal › Article › peer-review

19 Citations (Scopus)

Abstract

Existing aspect-based/category sentiment analysis methods have shown great success in detecting sentiment polarity toward a given aspect in a sentence with supervised learning, where the training and inference stages share the same pre-defined set of aspects. However, in practice, the aspect categories are changing rather than keeping fixed over time. Dealing with unseen aspect categories is under-explored in existing methods. In this article, we formulate a new few-shot aspect category sentiment analysis (FSACSA) task, which aims to effectively predict the sentiment polarity of previously unseen aspect categories. To this end, we propose a novel Aspect-Focused Meta-Learning (AFML) framework that constructs aspect-Aware and aspect-contrastive representations from external knowledge to match the target aspect with aspects in the training set. Concretely, we first construct two auxiliary contrastive sentences for a given sentence with the incorporation of external knowledge, enabling the learning of sentence representations with a better generalization. Then, we devise an aspect-focused induction network to leverage the contextual sentiment toward a given aspect to refine the label vectors. Furthermore, we employ the episode-based meta-learning algorithm to train the whole network, so as to learn to generalize to novel aspects. Extensive experiments on multiple real-life datasets show that our proposed AFML framework achieves the state-of-The-Art results for the FSACSA task.

Original language	English
Article number	22
Number of pages	31
Journal	ACM TRANSACTIONS ON INFORMATION SYSTEMS
Volume	41
Issue number	1
Early online date	22 Apr 2022
DOIs	https://doi.org/10.1145/3529954
Publication status	Published - 31 Jan 2023

Keywords

Few-shot aspect category sentiment analysis
few-shot learning
meta-learning
sentiment analysis

Access to Document

10.1145/3529954

Cite this

@article{03e90ef1799640488c1e1ae634496a1a,

title = "Few-shot Aspect Category Sentiment Analysis via Meta-learning",

abstract = "Existing aspect-based/category sentiment analysis methods have shown great success in detecting sentiment polarity toward a given aspect in a sentence with supervised learning, where the training and inference stages share the same pre-defined set of aspects. However, in practice, the aspect categories are changing rather than keeping fixed over time. Dealing with unseen aspect categories is under-explored in existing methods. In this article, we formulate a new few-shot aspect category sentiment analysis (FSACSA) task, which aims to effectively predict the sentiment polarity of previously unseen aspect categories. To this end, we propose a novel Aspect-Focused Meta-Learning (AFML) framework that constructs aspect-Aware and aspect-contrastive representations from external knowledge to match the target aspect with aspects in the training set. Concretely, we first construct two auxiliary contrastive sentences for a given sentence with the incorporation of external knowledge, enabling the learning of sentence representations with a better generalization. Then, we devise an aspect-focused induction network to leverage the contextual sentiment toward a given aspect to refine the label vectors. Furthermore, we employ the episode-based meta-learning algorithm to train the whole network, so as to learn to generalize to novel aspects. Extensive experiments on multiple real-life datasets show that our proposed AFML framework achieves the state-of-The-Art results for the FSACSA task.",

keywords = "Few-shot aspect category sentiment analysis, few-shot learning, meta-learning, sentiment analysis",

author = "Bin Liang and Xiang Li and Lin Gui and Yonghao Fu and Yulan He and Min Yang and Ruifeng Xu",

note = "Funding Information: This work was partially supported by the National Natural Science Foundation of China (61876053, 62006062, 62176076, 61906185), the Shenzhen Foundational Research Funding (JCYJ20200109113441941, JCYJ20210324115614039), Shenzhen Science and Technology Program (JSGG20210802154400001), UK Engineering and Physical Sciences Research Council (grant no. EP/V048597/1, EP/T017112/1), the Shenzhen Science and Technology Innovation Program (Grant No. KQTD20190929172835662), and the Joint Lab of Lab of HITSZ and China Merchants Securities. Yulan He is supported by a Turing AI Fellowship funded by the UK Research and Innovation (grant no. EP/V020579/1). Publisher Copyright: {\textcopyright} 2023 Association for Computing Machinery.",

year = "2023",

month = jan,

day = "31",

doi = "10.1145/3529954",

language = "English",

volume = "41",

journal = "ACM TRANSACTIONS ON INFORMATION SYSTEMS",

issn = "1046-8188",

publisher = "Association for Computing Machinery (ACM)",

number = "1",

}

TY - JOUR

T1 - Few-shot Aspect Category Sentiment Analysis via Meta-learning

AU - Liang, Bin

AU - Li, Xiang

AU - Gui, Lin

AU - Fu, Yonghao

AU - He, Yulan

AU - Yang, Min

AU - Xu, Ruifeng

N1 - Funding Information: This work was partially supported by the National Natural Science Foundation of China (61876053, 62006062, 62176076, 61906185), the Shenzhen Foundational Research Funding (JCYJ20200109113441941, JCYJ20210324115614039), Shenzhen Science and Technology Program (JSGG20210802154400001), UK Engineering and Physical Sciences Research Council (grant no. EP/V048597/1, EP/T017112/1), the Shenzhen Science and Technology Innovation Program (Grant No. KQTD20190929172835662), and the Joint Lab of Lab of HITSZ and China Merchants Securities. Yulan He is supported by a Turing AI Fellowship funded by the UK Research and Innovation (grant no. EP/V020579/1). Publisher Copyright: © 2023 Association for Computing Machinery.

PY - 2023/1/31

Y1 - 2023/1/31

N2 - Existing aspect-based/category sentiment analysis methods have shown great success in detecting sentiment polarity toward a given aspect in a sentence with supervised learning, where the training and inference stages share the same pre-defined set of aspects. However, in practice, the aspect categories are changing rather than keeping fixed over time. Dealing with unseen aspect categories is under-explored in existing methods. In this article, we formulate a new few-shot aspect category sentiment analysis (FSACSA) task, which aims to effectively predict the sentiment polarity of previously unseen aspect categories. To this end, we propose a novel Aspect-Focused Meta-Learning (AFML) framework that constructs aspect-Aware and aspect-contrastive representations from external knowledge to match the target aspect with aspects in the training set. Concretely, we first construct two auxiliary contrastive sentences for a given sentence with the incorporation of external knowledge, enabling the learning of sentence representations with a better generalization. Then, we devise an aspect-focused induction network to leverage the contextual sentiment toward a given aspect to refine the label vectors. Furthermore, we employ the episode-based meta-learning algorithm to train the whole network, so as to learn to generalize to novel aspects. Extensive experiments on multiple real-life datasets show that our proposed AFML framework achieves the state-of-The-Art results for the FSACSA task.

AB - Existing aspect-based/category sentiment analysis methods have shown great success in detecting sentiment polarity toward a given aspect in a sentence with supervised learning, where the training and inference stages share the same pre-defined set of aspects. However, in practice, the aspect categories are changing rather than keeping fixed over time. Dealing with unseen aspect categories is under-explored in existing methods. In this article, we formulate a new few-shot aspect category sentiment analysis (FSACSA) task, which aims to effectively predict the sentiment polarity of previously unseen aspect categories. To this end, we propose a novel Aspect-Focused Meta-Learning (AFML) framework that constructs aspect-Aware and aspect-contrastive representations from external knowledge to match the target aspect with aspects in the training set. Concretely, we first construct two auxiliary contrastive sentences for a given sentence with the incorporation of external knowledge, enabling the learning of sentence representations with a better generalization. Then, we devise an aspect-focused induction network to leverage the contextual sentiment toward a given aspect to refine the label vectors. Furthermore, we employ the episode-based meta-learning algorithm to train the whole network, so as to learn to generalize to novel aspects. Extensive experiments on multiple real-life datasets show that our proposed AFML framework achieves the state-of-The-Art results for the FSACSA task.

KW - Few-shot aspect category sentiment analysis

KW - few-shot learning

KW - meta-learning

KW - sentiment analysis

UR - http://www.scopus.com/inward/record.url?scp=85143846850&partnerID=8YFLogxK

U2 - 10.1145/3529954

DO - 10.1145/3529954

M3 - Article

AN - SCOPUS:85143846850

SN - 1046-8188

VL - 41

JO - ACM TRANSACTIONS ON INFORMATION SYSTEMS

JF - ACM TRANSACTIONS ON INFORMATION SYSTEMS

IS - 1

M1 - 22

ER -

Few-shot Aspect Category Sentiment Analysis via Meta-learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this