西华大学学报（自然科学版）2026，Vol.45Issue(2)：74-83,10.DOI:10.12198/j.issn.1673-159X.5565
方面级情感分析的知识增强提示模板构建
Construction of Knowledge-Enhancing Prompt Templates for Aspect-Level Sentiment Analysis
摘要
Abstract
Currently,aspect-level sentiment analysis models based on neural networks primarily in-volve discrete training of two subtasks:aspect term extraction and aspect polarity classification,or fine-tun-ing of pre-trained language models.These approaches neglect the interplay between the two subtasks,mak-ing it challenging to fully leverage the language model knowledge acquired during pre-training.This paper proposes an external knowledge-enhanced prompt template approach(KPT),which constructs the vocabu-lary of the pre-trained language model into a K-dimensional tree.The K-nearest neighbor search algorithm is employed to search for optimal prompt words on this K-dimensional tree,thereby constructing an optim-al prompt template.In this process,external knowledge is integrated into the vocabulary to enrich its se-mantic information.Features from the aspect term extraction task are utilized to further enhance the per-formance of the aspect polarity classification task.A multi-head attention mechanism is employed to enable interaction between the two tasks and integrate them,thereby improving the accuracy of sentiment polarity judgment by the external knowledge-enhanced prompt template.Experimental results on three public data-sets,namely Lap14,Rest14,and Twitter,demonstrate that the proposed method outperforms existing mod-els such as ASGCN,BiGCN,CDT,and others.关键词
方面级情感分析/外部知识/提示模板Key words
aspect-level sentiment analysis/external knowledge/prompt template分类
信息技术与安全科学引用本文复制引用
张茂琳,李显勇,杜亚军,黄东..方面级情感分析的知识增强提示模板构建[J].西华大学学报（自然科学版）,2026,45(2):74-83,10.基金项目
宜宾市科技计划项目(2023SF004) （2023SF004）
四川省科技计划项目(2022YFG0378、2023YFS0424、2023YFH0058、2023YFQ0044) （2022YFG0378、2023YFS0424、2023YFH0058、2023YFQ0044）
非遗数字化与多源信息融合福建省高校工程研究中心(福建技术师范学院)项目(G3-KF2022). （福建技术师范学院）
