A debiased self-training framework with graph self-supervised pre-training aided for semi-supervised rumor detection

计算机科学 谣言 杠杆(统计) 人工智能 机器学习 图形 标记数据 半监督学习 监督学习 训练集 模式识别(心理学) 人工神经网络 公共关系 理论计算机科学 政治学
作者
Yuhan Qiao,Chaoqun Cui,Yiying Wang,Caiyan Jia
出处
期刊:Neurocomputing [Elsevier BV]
卷期号:604: 128314-128314
标识
DOI:10.1016/j.neucom.2024.128314
摘要

Existing rumor detection models have achieved remarkable performance in fully-supervised settings. However, it is time-consuming and labor-intensive to obtain extensive labeled rumor data. To mitigate the reliance on labeled data, semi-supervised learning (SSL), jointly learning from labeled and unlabeled samples, achieves significant performance improvements at low costs. Commonly used self-training methods in SSL, despite their simplicity and efficiency, suffer from the notorious confirmation bias, which can be seen as the accumulation of noise arising from utilization of incorrect pseudo-labels. To deal with the problem, in this study, we propose a debiased self-training framework with graph self-supervised pre-training for semi-supervised rumor detection. First, to enhance the initial model for self-training and reduce the generation of incorrect pseudo-labels in early stages, we leverage the rumor propagation structures of massive unlabeled data for graph self-supervised pre-training. Second, we improve the quality of pseudo-labels by proposing a pseudo-labeling strategy with self-adaptive thresholds, which consists of self-paced global thresholds controlling the overall utilization process of pseudo-labels and local class-specific thresholds attending to the learning status of each class. Extensive experiments on four public benchmarks demonstrate that our method significantly outperforms previous rumor detection baselines in semi-supervised settings, especially when labeled samples are extremely scarce. Notably, we have achieved 96.3% accuracy on Weibo with 500 labels per class and 86.0% accuracy with just 5 labels per class.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

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
siestaMiao发布了新的文献求助10
刚刚
a毛发布了新的文献求助10
1秒前
1秒前
华东小可爱完成签到,获得积分10
2秒前
3秒前
啦啦啦发布了新的文献求助10
3秒前
Hello应助齐司礼的小笨鸟采纳,获得10
5秒前
6秒前
xuan完成签到,获得积分10
6秒前
tjfwg完成签到,获得积分10
7秒前
Jean完成签到,获得积分10
9秒前
星光熠熠完成签到 ,获得积分20
10秒前
小智完成签到 ,获得积分10
12秒前
12秒前
13秒前
14秒前
15秒前
15秒前
馨馨完成签到,获得积分20
16秒前
雪山飞龙发布了新的文献求助10
16秒前
kehan发布了新的文献求助10
17秒前
18秒前
小智关注了科研通微信公众号
18秒前
无花果应助激动的一手采纳,获得10
19秒前
dd99081完成签到,获得积分10
19秒前
谨慎初曼发布了新的文献求助10
19秒前
happy发布了新的文献求助10
21秒前
爱科研的小曹完成签到,获得积分10
21秒前
21秒前
心灵的守望完成签到,获得积分10
22秒前
22秒前
23秒前
八块腹肌完成签到 ,获得积分10
24秒前
25秒前
siestaMiao完成签到,获得积分10
25秒前
JamesPei应助谨慎初曼采纳,获得10
26秒前
whb发布了新的文献求助20
27秒前
27秒前
怕黑海冬发布了新的文献求助50
27秒前
fox2shj完成签到,获得积分10
28秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
‘Unruly’ Children: Historical Fieldnotes and Learning Morality in a Taiwan Village (New Departures in Anthropology) 400
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
基于可调谐半导体激光吸收光谱技术泄漏气体检测系统的研究 330
Composite Predicates in English 300
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3982367
求助须知:如何正确求助?哪些是违规求助? 3526007
关于积分的说明 11229870
捐赠科研通 3263850
什么是DOI,文献DOI怎么找? 1801703
邀请新用户注册赠送积分活动 879994
科研通“疑难数据库(出版商)”最低求助积分说明 807767