@inproceedings{zhao-etal-2022-kesa,
title = "{KESA}: A Knowledge Enhanced Approach To Sentiment Analysis",
author = "Zhao, Qinghua and
Ma, Shuai and
Ren, Shuo",
booktitle = "Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = nov,
year = "2022",
address = "Online only",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.aacl-main.58",
pages = "766--776",
abstract = "Though some recent works focus on injecting sentiment knowledge into pre-trained language models, they usually design mask and reconstruction tasks in the post-training phase. This paper aims to integrate sentiment knowledge in the fine-tuning stage. To achieve this goal, we propose two sentiment-aware auxiliary tasks named sentiment word selection and conditional sentiment prediction and, correspondingly, integrate them into the objective of the downstream task. The first task learns to select the correct sentiment words from the given options. The second task predicts the overall sentiment polarity, with the sentiment polarity of the word given as prior knowledge. In addition, two label combination methods are investigated to unify multiple types of labels in each auxiliary task. Experimental results demonstrate that our approach consistently outperforms baselines (achieving a new state-of-the-art) and is complementary to existing sentiment-enhanced post-trained models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2022-kesa">
<titleInfo>
<title>KESA: A Knowledge Enhanced Approach To Sentiment Analysis</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qinghua</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuai</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuo</namePart>
<namePart type="family">Ren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online only</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Though some recent works focus on injecting sentiment knowledge into pre-trained language models, they usually design mask and reconstruction tasks in the post-training phase. This paper aims to integrate sentiment knowledge in the fine-tuning stage. To achieve this goal, we propose two sentiment-aware auxiliary tasks named sentiment word selection and conditional sentiment prediction and, correspondingly, integrate them into the objective of the downstream task. The first task learns to select the correct sentiment words from the given options. The second task predicts the overall sentiment polarity, with the sentiment polarity of the word given as prior knowledge. In addition, two label combination methods are investigated to unify multiple types of labels in each auxiliary task. Experimental results demonstrate that our approach consistently outperforms baselines (achieving a new state-of-the-art) and is complementary to existing sentiment-enhanced post-trained models.</abstract>
<identifier type="citekey">zhao-etal-2022-kesa</identifier>
<location>
<url>https://aclanthology.org/2022.aacl-main.58</url>
</location>
<part>
<date>2022-11</date>
<extent unit="page">
<start>766</start>
<end>776</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T KESA: A Knowledge Enhanced Approach To Sentiment Analysis
%A Zhao, Qinghua
%A Ma, Shuai
%A Ren, Shuo
%S Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2022
%8 November
%I Association for Computational Linguistics
%C Online only
%F zhao-etal-2022-kesa
%X Though some recent works focus on injecting sentiment knowledge into pre-trained language models, they usually design mask and reconstruction tasks in the post-training phase. This paper aims to integrate sentiment knowledge in the fine-tuning stage. To achieve this goal, we propose two sentiment-aware auxiliary tasks named sentiment word selection and conditional sentiment prediction and, correspondingly, integrate them into the objective of the downstream task. The first task learns to select the correct sentiment words from the given options. The second task predicts the overall sentiment polarity, with the sentiment polarity of the word given as prior knowledge. In addition, two label combination methods are investigated to unify multiple types of labels in each auxiliary task. Experimental results demonstrate that our approach consistently outperforms baselines (achieving a new state-of-the-art) and is complementary to existing sentiment-enhanced post-trained models.
%U https://aclanthology.org/2022.aacl-main.58
%P 766-776
Markdown (Informal)
[KESA: A Knowledge Enhanced Approach To Sentiment Analysis](https://aclanthology.org/2022.aacl-main.58) (Zhao et al., AACL-IJCNLP 2022)
ACL
- Qinghua Zhao, Shuai Ma, and Shuo Ren. 2022. KESA: A Knowledge Enhanced Approach To Sentiment Analysis. In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 766–776, Online only. Association for Computational Linguistics.