亲爱的研友该休息了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!身体可是革命的本钱,早点休息,好梦!

Semantic Consistency Regularization with Large Language Models for Semi-supervised Sentiment Analysis

正规化(语言学) 一致性(知识库) 自然语言处理 计算机科学 人工智能 情绪分析
作者
Kunrong Li,Xinyu Liu,Zhen Chen
出处
期刊:Cornell University - arXiv [Cornell University]
标识
DOI:10.48550/arxiv.2501.17598
摘要

Accurate sentiment analysis of texts is crucial for a variety of applications, such as understanding customer feedback, monitoring market trends, and detecting public sentiment. However, manually annotating large sentiment corpora for supervised learning is labor-intensive and time-consuming. Therefore, it is essential and effective to develop a semi-supervised method for the sentiment analysis task. Although some methods have been proposed for semi-supervised text classification, they rely on the intrinsic information within the unlabeled data and the learning capability of the NLP model, which lack generalization ability to the sentiment analysis scenario and may prone to overfit. Inspired by the ability of pretrained Large Language Models (LLMs) in following instructions and generating coherent text, we propose a Semantic Consistency Regularization with Large Language Models (SCR) framework for semi-supervised sentiment analysis. We introduce two prompting strategies to semantically enhance unlabeled text using LLMs. The first is Entity-based Enhancement (SCR-EE), which involves extracting entities and numerical information, and querying the LLM to reconstruct the textual information. The second is Concept-based Enhancement (SCR-CE), which directly queries the LLM with the original sentence for semantic reconstruction. Subsequently, the LLM-augmented data is utilized for a consistency loss with confidence thresholding, which preserves high-quality agreement samples to provide additional supervision signals during training. Furthermore, to fully utilize the uncertain unlabeled data samples, we propose a class re-assembling strategy inspired by the class space shrinking theorem. Experiments show our method achieves remarkable performance over prior semi-supervised methods.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
jundongfan完成签到,获得积分20
2秒前
4秒前
wanci应助Jason采纳,获得20
5秒前
6秒前
jjj完成签到,获得积分10
8秒前
su完成签到,获得积分20
8秒前
1123发布了新的文献求助10
8秒前
8秒前
10秒前
su发布了新的文献求助10
11秒前
12秒前
石中酒发布了新的文献求助10
13秒前
清爽的珍发布了新的文献求助10
14秒前
胖大海完成签到 ,获得积分10
15秒前
黄焖鸡米饭完成签到 ,获得积分10
18秒前
20秒前
大力牌皮揣子完成签到 ,获得积分10
21秒前
充电宝应助石中酒采纳,获得10
21秒前
852应助石中酒采纳,获得10
21秒前
Ava应助石中酒采纳,获得20
21秒前
慕青应助石中酒采纳,获得10
21秒前
大模型应助石中酒采纳,获得20
22秒前
Jasper应助石中酒采纳,获得20
22秒前
大爱仙尊发布了新的文献求助10
28秒前
wyh完成签到,获得积分20
29秒前
29秒前
30秒前
李爱国应助科研通管家采纳,获得10
31秒前
31秒前
小蘑菇应助科研通管家采纳,获得10
31秒前
852应助清爽的珍采纳,获得10
32秒前
33完成签到,获得积分10
32秒前
Malik发布了新的文献求助10
38秒前
石中酒完成签到,获得积分10
38秒前
ZJ发布了新的文献求助10
41秒前
43秒前
上官若男应助Malik采纳,获得10
43秒前
CyberHamster完成签到,获得积分0
44秒前
跳跃的访曼完成签到,获得积分10
47秒前
高分求助中
Principles of Economics, 11th Edition 10000
University Physics with Modern Physics, 16th edition 10000
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
48V Low-voltage Power Distribution Network (PDN) Architecture Industry Report, 2024 800
ズームレンズの光学設計に関する研究 800
Fundamentals of Pharmaceutical and Biologics Regulations: A Global Perspective, Second Edition 700
Matrix Methods in Data Mining and Pattern Recognition Second Edition 610
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 工程类 有机化学 化学工程 生物化学 计算机科学 内科学 物理 复合材料 催化作用 细胞生物学 无机化学 光电子学 物理化学 电极 基因
热门帖子
关注 科研通微信公众号,转发送积分 7297287
求助须知:如何正确求助?哪些是违规求助? 8915741
关于积分的说明 18878850
捐赠科研通 6963004
什么是DOI,文献DOI怎么找? 3210524
关于科研通互助平台的介绍 2379855
邀请新用户注册赠送积分活动 2187016