Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis

This study proposes an Aspect-Enhanced Prompting (AEP) method for unsupervised Multi-Source Domain Adaptation in Aspect Sentiment Classification, where data from the target domain are completely unavailable for model training. The proposed AEP is based on two generative language models: one generate...

Full description

Saved in:
Bibliographic Details
Main Authors: Binghan Lu, Kiyoaki Shirai, Natthawut Kertkeidkachorn
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Information
Subjects:
Online Access:https://www.mdpi.com/2078-2489/16/5/411
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327044957044736
author Binghan Lu
Kiyoaki Shirai
Natthawut Kertkeidkachorn
author_facet Binghan Lu
Kiyoaki Shirai
Natthawut Kertkeidkachorn
author_sort Binghan Lu
collection DOAJ
description This study proposes an Aspect-Enhanced Prompting (AEP) method for unsupervised Multi-Source Domain Adaptation in Aspect Sentiment Classification, where data from the target domain are completely unavailable for model training. The proposed AEP is based on two generative language models: one generates a prompt from a given review, while the other follows the prompt and classifies the sentiment of an aspect. The first model extracts Aspect-Related Features (ARFs), which are words closely related to the aspect, from the review and incorporates them into the prompt in a domain-agnostic manner, thereby directing the second model to identify the sentiment accurately. Our framework incorporates an innovative rescoring mechanism and a cluster-based prompt expansion strategy. Both are intended to enhance the robustness of the generation of the prompt and the adaptability of the model to diverse domains. The results of experiments conducted on five datasets (Restaurant, Laptop, Device, Service, and Location) demonstrate that our method outperforms the baselines, including a state-of-the-art unsupervised domain adaptation method. The effectiveness of both the rescoring mechanism and the cluster-based prompt expansion is also validated through an ablation study.
format Article
id doaj-art-f30a5167237345a19a529d4dff52f3d8
institution Kabale University
issn 2078-2489
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Information
spelling doaj-art-f30a5167237345a19a529d4dff52f3d82025-08-20T03:47:59ZengMDPI AGInformation2078-24892025-05-0116541110.3390/info16050411Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment AnalysisBinghan Lu0Kiyoaki Shirai1Natthawut Kertkeidkachorn2Graduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa 923-1292, JapanGraduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa 923-1292, JapanGraduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa 923-1292, JapanThis study proposes an Aspect-Enhanced Prompting (AEP) method for unsupervised Multi-Source Domain Adaptation in Aspect Sentiment Classification, where data from the target domain are completely unavailable for model training. The proposed AEP is based on two generative language models: one generates a prompt from a given review, while the other follows the prompt and classifies the sentiment of an aspect. The first model extracts Aspect-Related Features (ARFs), which are words closely related to the aspect, from the review and incorporates them into the prompt in a domain-agnostic manner, thereby directing the second model to identify the sentiment accurately. Our framework incorporates an innovative rescoring mechanism and a cluster-based prompt expansion strategy. Both are intended to enhance the robustness of the generation of the prompt and the adaptability of the model to diverse domains. The results of experiments conducted on five datasets (Restaurant, Laptop, Device, Service, and Location) demonstrate that our method outperforms the baselines, including a state-of-the-art unsupervised domain adaptation method. The effectiveness of both the rescoring mechanism and the cluster-based prompt expansion is also validated through an ablation study.https://www.mdpi.com/2078-2489/16/5/411aspect-based sentiment analysisunsupervised domain adaptationmulti-source domain adaptationtext generation modelprompt engineering
spellingShingle Binghan Lu
Kiyoaki Shirai
Natthawut Kertkeidkachorn
Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
Information
aspect-based sentiment analysis
unsupervised domain adaptation
multi-source domain adaptation
text generation model
prompt engineering
title Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
title_full Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
title_fullStr Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
title_full_unstemmed Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
title_short Aspect-Enhanced Prompting Method for Unsupervised Domain Adaptation in Aspect-Based Sentiment Analysis
title_sort aspect enhanced prompting method for unsupervised domain adaptation in aspect based sentiment analysis
topic aspect-based sentiment analysis
unsupervised domain adaptation
multi-source domain adaptation
text generation model
prompt engineering
url https://www.mdpi.com/2078-2489/16/5/411
work_keys_str_mv AT binghanlu aspectenhancedpromptingmethodforunsuperviseddomainadaptationinaspectbasedsentimentanalysis
AT kiyoakishirai aspectenhancedpromptingmethodforunsuperviseddomainadaptationinaspectbasedsentimentanalysis
AT natthawutkertkeidkachorn aspectenhancedpromptingmethodforunsuperviseddomainadaptationinaspectbasedsentimentanalysis