Improving Cyberbullying Detection with User Interaction
Improving Cyberbullying Detection with User Interaction
复制标题
DOI：
10.1145/3442381.3449828
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Suyu Ge;Lu Cheng;Huan Liu
中科院分区：
文献类型：
--
作者：
Suyu Ge;Lu Cheng;Huan Liu
网络欺凌被认为是有意且重复的网络欺凌行为，在过去几十年中变得越来越普遍。尽管迄今为止取得了重大进展，但大多数现有网络欺凌检测工作的重点在于对社交媒体会话中不同评论的独立内容分析。我们认为，这种主要的分析概念存在三个关键局限性：它们忽视了不同评论之间的时间相关性；他们只考虑单个评论中的内容，而不考虑评论之间的主题连贯性；它们仍然是通用的，并且利用社交媒体用户之间有限的互动。在这项工作中，我们观察到同一会话中的用户评论可能本质上相关，例如讨论相似的主题，并且他们的交互可能随着时间的推移而演变。我们还表明，对此类主题连贯性和时间交互进行建模对于捕获欺凌行为的重复特征至关重要，从而获得更好的预测性能。为了实现这一目标，我们首先为每个社交媒体会话构建一个统一的时间图。利用图神经网络的最新进展，我们提出了一种基于图的原理方法，用于对整个用户交互过程中的时间动态和主题连贯性进行建模。我们通过会话级欺凌检测和评论级案例研究的任务来实证评估我们方法的有效性。我们的代码向公众发布。 1
Cyberbullying, identified as intended and repeated online bullying behavior, has become increasingly prevalent in the past few decades. Despite the significant progress made thus far, the focus of most existing work on cyberbullying detection lies in the independent content analysis of different comments within a social media session. We argue that such leading notions of analysis suffer from three key limitations: they overlook the temporal correlations among different comments; they only consider the content within a single comment rather than the topic coherence across comments; they remain generic and exploit limited interactions between social media users. In this work, we observe that user comments in the same session may be inherently related, e.g., discussing similar topics, and their interaction may evolve over time. We also show that modeling such topic coherence and temporal interaction are critical to capture the repetitive characteristics of bullying behavior, thus leading to better predicting performance. To achieve the goal, we first construct a unified temporal graph for each social media session. Drawing on recent advances in graph neural network, we then propose a principled graph-based approach for modeling the temporal dynamics and topic coherence throughout user interactions. We empirically evaluate the effectiveness of our approach with the tasks of session-level bullying detection and comment-level case study. Our code is released to public. 1