分析社交媒体中仇恨与反仇恨言论的主题和修辞特征
Dialogues of Dissent: Thematic and Rhetorical Dimensions of Hate and Counter-Hate Speech in Social Media Conversations
- 构建多标签体系，从主题与修辞双维度标注仇恨与反仇恨言论
- 分析720条推文，发现仇恨言论常以情感煽动为主，反仇恨言论更依赖逻辑论证
- 揭示仇恨与反仇恨言论的互动模式，适合舆情分析与平台治理研究者
我们提出一种新型多标签标注方案，用于联合标注社交媒体对话中的仇恨与反仇恨言论，将其划分为主题与修辞两个维度。主题类别涵盖各类言论的论述层面，修辞维度则基于亚里士多德的逻各斯、埃托斯与帕托斯理论，捕捉言论表达方式。我们标注了92组对话共720条推文，并结合公开数据指标进行统计分析，探究仇恨与反仇恨言论在主题与修辞维度上的内部及相互作用模式。研究结果揭示了仇恨言论在社交平台上的传播机制、反制策略及其对网络行为的潜在影响。
原文摘要 · Abstract (English)
We introduce a novel multi-labeled scheme for joint annotation of hate and counter-hate speech in social media conversations, categorizing hate and counter-hate messages into thematic and rhetorical dimensions. The thematic categories outline different discursive aspects of each type of speech, while the rhetorical dimension captures how hate and counter messages are communicated, drawing on Aristotle's Logos, Ethos and Pathos. We annotate a sample of 92 conversations, consisting of 720 tweets, and conduct statistical analyses, incorporating public metrics, to explore patterns of interaction between the thematic and rhetorical dimensions within and between hate and counter-hate speech. Our findings provide insights into the spread of hate messages on social media, the strategies used to counter them, and their potential impact on online behavior.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。