谣言检测(ClaHi-GAT)《Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks》
论文信息
论文标题:Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks
论文作者:Erxue Min, Yu Rong, Yatao Bian, Tingyang Xu, Peilin Zhao, Junzhou Huang,Sophia Ananiadou
论文来源:2021,EMNLP
论文地址:download
论文代码:download
Background
传播结构为谣言的真假提供了有用的线索,但是现有的谣言检测方法要么局限于用户相应关系,要么简化了对话结构。
本文说的 Claim 代表的是 Source post ,即源帖。
1 Introduction
如下为一个简单的 conversation thread 例子:

本文提出的点:考虑兄弟之间的关系,如下图虚线部分。

2 Claim-guided Hierarchical Graph Attention Networks
总体框架如下:

本文的模型包括两个注意力模块:
- A Graph Attention to capture the importance of different neighboring tweets
- A claim-guided hierarchical attention to enhance post content understanding
2.1 Claim-guided Hierarchical Attention
对于每个 tweet $x_i$ ,首先使用 Bi-LSTM 获得 Post 的特征矩阵 $X=\left[c, x_{1}, x_{2}, \cdots, x_{|\mathcal{V}|-1}\right]^{\top}$ ,其中 $c, x_{i} \in \mathbb{R}^{d}$。
为加强模型的主题一致性和语义推理:
Post-level Attention
为了防止主题偏离和丢失 claim 的信息,本文采用 gate module 决定它应该接受 claim 多少信息,以更好地指导相关职位的重要性分配。claim-aware representation 具体如下:
$\begin{array}{l}g_{c \rightarrow x_{i}}^{(l)} &=&\operatorname{sigmoid}\left(W_{g}^{(l)} h_{x_{i}}^{(l)}+U_{g}^{(l)} h_{c}^{(l)}\right) \\\tilde{h}_{x_{i}}^{(l)} &=&g_{c \rightarrow x_{i}}^{(l)} \odot h_{x_{i}}^{(l)}+\left(1-g_{c \rightarrow x_{i}}^{(l)}\right) \odot h_{c}^{(l)}\end{array}$
其中,$g_{c \rightarrow x_{i}}^{(l)}$ 是一个 gate vector,$W_{g}^{(l)}$ 和 $U_{g}^{(l)}$ 是可学习参数。
然后,将 claim-aware representation 与 original representation 拼接起来,作为 $\text{Eq.1}$ 的输入去计算注意力权重:
$\begin{array}{l}\hat{h}_{x_{i}}^{(l)}=\left[\tilde{h}_{x_{i}}^{(l)} \| h_{x_{i}}^{(l)}\right] \\\hat{\alpha}_{i, j}^{(l)}=\operatorname{Atten}\left(\hat{h}_{x_{i}}^{(l)}, \hat{h}_{x_{j}}^{(l)}\right)\end{array}$
2.2 Graph Attention Networks
为了编码结构信息,本文使用 GAT encoder:
输入:$H^{(l)}=\left[h_{c}^{(l)}, h_{x_{1}}^{(l)}, h_{x_{2}}^{(l)}, \ldots, h_{x_{|\mathcal{V}|-1}}^{(l)}\right]^{\top}$
过程:
${\large \begin{aligned}\alpha_{i, j}^{(l)} &=\operatorname{Atten}\left(h_{x_{i}}^{(l)}, h_{x_{j}}^{(l)}\right) \\&=\frac{\exp \left(\phi\left(a^{\top}\left[W^{(l)} h_{x_{i}}^{(l)} \| W^{(l)} h_{x_{j}}^{(l)}\right]\right)\right)}{\sum_{j \in \mathcal{N}_{i}} \exp \left(\phi\left(a^{\top}\left[W^{(l)} h_{x_{i}}^{(l)} \| W^{(l)} h_{x_{j}}^{(l)}\right]\right)\right)}\end{aligned}} $
$h_{x_{i}}^{(l+1)}=\operatorname{Re} L U\left(\sum\limits_{j \in \mathcal{N}_{i}} \alpha_{i, j}^{(l)} W^{(l)} h_{x_{j}}^{(l)}\right)$
考虑多头注意力:
$h_{x_{i}}^{(l+1)}=\|_{k=1}^{K} \operatorname{ReLU}\left(\sum\limits _{j \in \mathcal{N}_{i}} \alpha_{i, j}^{(l, k)} W_{k}^{(l)} h_{x_{j}}^{(l)}\right)$
替换输出层的表示向量:
${\large h_{x_{i}}^{(L)}=\operatorname{Re} L U\left(\frac{1}{K} \sum\limits _{k=1}^{K} \sum\limits_{j \in \mathcal{N}_{i}} \alpha_{i, j}^{\left(l^{\prime}, k\right)} W_{k}^{\left(l^{\prime}\right)} h_{x_{j}}^{\left(l^{\prime}\right)}\right)} $
输出:图表示
$\bar{s}=\text { mean-pooling }\left(H^{(L)}\right)$
Event-level Attention
出发点:获得图表示的时候采用的 平均池化并不是一定有意义的,可能存在某些节点对于图分类来说更准确。
受到 Natural Language Inference (NLI) 的影响,本文考虑对 GAT 最后一层的 $h_{c}^{(L)}$ 和 $\left.h_{x_{i}}^{(L)}: 1\right)$ 做如下处理 :
1)concatenation $\left[h_{c}^{(L)} \| h_{x_{i}}^{(L)}\right]$
2)element-wise product $h_{\text {prod }}^{(L)}=h_{c}^{(L)} \odot h_{x_{i}}^{(L)}$
3)absolute element-wise difference $h_{\text {diff }}^{(L)}=\left|h_{c}^{(L)}-h_{x_{i}}^{(L)}\right|$
接着获得一个联合表示:
$h_{x_{i}}^{c}=\tanh \left(F C\left(\left[h_{c}^{(L)}\left\|h_{x_{i}}^{(L)}\right\| h_{\text {prod }}^{(L)} \| h_{\text {diff }}^{(L)}\right]\right)\right)$
通过使用该联合表示计算 Event-level Attention :
${\large \begin{array}{l}b_{i} &=&\tanh \left(F C\left(h_{x_{i}}^{c}\right)\right) \\\beta_{i} &=&\frac{\exp \left(b_{i}\right)}{\sum_{i} \exp \left(b_{i}\right)} \\\hat{s} &&=\sum_{i} \beta_{i} h_{x_{i}}^{(L)}\end{array}} $
最后将其 $\hat{S}$ 与 GAT 最后一层的平均池化图表示 $\bar{s}$ 拼接作为最终图表示,并进行分类:
$\hat{y}=\operatorname{softmax}(F C([\hat{s} \| \bar{s}]))$
3 Experiments

3.2 Rumor Classifification Performance


3.3 Ablation Study
1) ClaHi-GAT/DT: Instead of the undirected interaction graph, we use the directed trees as the model input.
2) GAT+EA+SC: We simply concatenate the features of the claim with the node features at each GAT layer, to replace the claim-aware representation.
3) w/o EA: We discard the event-level (inference-based) attention as presented.
4) w/o PA: We neglect the post-level (claim-aware) attention by leaving out the gating module introduced.
5) GAT: The backbone model.
6) GCN: The vanilla graph convolutional networks with no attention.

3.4 Evaluation of Undirected Interaction Graphs
- ClaHi-GAT/DT Utilize the directional tree applied in past influential works as the modeling way instead of our proposed undirected interaction graph.
- ClaHi-GAT/DTS Based on the directional tree structure similar to ClaHi-GAT/DT but the explicit interactions between sibling nodes are taken into account.
- ClaHi-GAT/UD The modeling way is our undirected interaction topology but without considering the explicit correlations between sibling nodes that reply to the same target.
- ClaHi-GAT In this paper, we propose to model the conversation thread as an undirected interaction graph for our claim-guided hierarchical graph attention networks.

3.5 Early Rumor Detection

关键点:随着 claim 的传播,或多或少会产生更多的语义信息和噪声,所以使用 claim 的信息至关重要。
举例说明:false claim 的注意力分数得分图如下:

言下之意:错误的 post $x_2$ 会被赋予较小的权重,这就是为什么该模型早期谣言检测比较稳定的原因。
谣言检测(ClaHi-GAT)《Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks》的更多相关文章
- 论文解读(RvNN)《Rumor Detection on Twitter with Tree-structured Recursive Neural Networks》
论文信息 论文标题:Rumor Detection on Twitter with Tree-structured Recursive Neural Networks论文作者:Jing Ma, Wei ...
- 谣言检测(PSIN)——《Divide-and-Conquer: Post-User Interaction Network for Fake News Detection on Social Media》
论文信息 论文标题:Divide-and-Conquer: Post-User Interaction Network for Fake News Detection on Social Media论 ...
- 谣言检测——《MFAN: Multi-modal Feature-enhanced Attention Networks for Rumor Detection》
论文信息 论文标题:MFAN: Multi-modal Feature-enhanced Attention Networks for Rumor Detection论文作者:Jiaqi Zheng, ...
- 谣言检测——(PSA)《Probing Spurious Correlations in Popular Event-Based Rumor Detection Benchmarks》
论文信息 论文标题:Probing Spurious Correlations in Popular Event-Based Rumor Detection Benchmarks论文作者:Jiayin ...
- 谣言检测(GACL)《Rumor Detection on Social Media with Graph Adversarial Contrastive Learning》
论文信息 论文标题:Rumor Detection on Social Media with Graph AdversarialContrastive Learning论文作者:Tiening Sun ...
- 谣言检测(PLAN)——《Interpretable Rumor Detection in Microblogs by Attending to User Interactions》
论文信息 论文标题:Interpretable Rumor Detection in Microblogs by Attending to User Interactions论文作者:Ling Min ...
- 谣言检测(RDEA)《Rumor Detection on Social Media with Event Augmentations》
论文信息 论文标题:Rumor Detection on Social Media with Event Augmentations论文作者:Zhenyu He, Ce Li, Fan Zhou, Y ...
- 谣言检测()《Data Fusion Oriented Graph Convolution Network Model for Rumor Detection》
论文信息 论文标题:Data Fusion Oriented Graph Convolution Network Model for Rumor Detection论文作者:Erxue Min, Yu ...
- 谣言检测()《Rumor Detection with Self-supervised Learning on Texts and Social Graph》
论文信息 论文标题:Rumor Detection with Self-supervised Learning on Texts and Social Graph论文作者:Yuan Gao, Xian ...
随机推荐
- PHP切割汉字
<?php /* @UTF-8编码的字符可能由1~3个字节组成. */ /*--------------------------方法一截取中文字符串方法--------------------- ...
- 中国联通改造 Apache DolphinScheduler 资源中心,实现计费环境跨集群调用与数据脚本一站式访问
截止2022年,中国联通用户规模达到4.6亿,占据了全中国人口的30%,随着5G的推广普及,运营商IT系统普遍面临着海量用户.海量话单.多样化业务.组网模式等一系列变革的冲击. 当前,联通每天处理话单 ...
- js中数组去重的方法
在实际工作或面试中,我们经常会遇到"数组去重"问题,接下来就是使用js实现的数组去重的多种方法: 1.借助ES6提供的Set结构 var arr = [1,1,2,2,3,3,4, ...
- React报错之Encountered two children with the same key
正文从这开始~ 总览 当我们从map()方法返回的两个或两个以上的元素具有相同的key属性时,会产生"Encountered two children with the same key&q ...
- java-Servlet编码/异常处理
1. Servlet输出中文(1)为什么会有乱码?out.println方法在输出时或者表单提交的时候,浏览器会对表单中的中文参数值进行编码; 注:会使用表单所在的页面打开时使用的编码方式进行编码服务 ...
- App切换到后台后如何保持持续定位?
为了保护用户隐私,大多数应用只会在前台运行时获取用户位置,当应用在后台运行时,定位功能会被禁止.这就导致APP在后台或者锁屏时无法正常记录GPS轨迹,这对打车.共享出行.跑步等需要实时记录用户轨迹的应 ...
- JUC源码学习笔记5——线程池,FutureTask,Executor框架源码解析
JUC源码学习笔记5--线程池,FutureTask,Executor框架源码解析 源码基于JDK8 参考了美团技术博客 https://tech.meituan.com/2020/04/02/jav ...
- C++ 遍历磁盘文件 非递归方法 和递归方法
1: 非递归方法: 一起学习 寻找快乐 // File Name: CSearch.h #pragma once #include <vector> #include <atlst ...
- JavaScript基础回顾知识点记录1
js执行顺序为从上往下执行 js中有6种数据类型 基本数据类型为: String Number Boolean Null Undefined 引用数据类型为: Object 使用typeof 查看对象 ...
- Javascript之异步循环打印这道小题
这道题,我相信很多前端从业者都知道,它本质上来说并不复杂,但是却可以有很深远的扩展,最终核心的主题其实就是异步的遍历,其中对于题目的初级解法,还涉及到一些作用域的知识.那么我们以最容易理解的解法入手, ...