A (More) Realistic Evaluation Setup for Generalisation of Community Models on Malicious Content Detection

arXiv:2404.01822v1 Announce Type: new
Abstract: Community models for malicious content detection, which take into account the context from a social graph alongside the content itself, have shown remarkable performance on benchmark datasets. Yet, misinformation and hate speech continue to propagate on social media networks. This mismatch can be partially attributed to the limitations of current evaluation setups that neglect the rapid evolution of online content and the underlying social graph. In this paper, we propose a novel evaluation setup for …

abstract arxiv benchmark community context cs.cl cs.lg cs.si datasets detection evaluation graph hate speech malicious content media misinformation networks performance setup social social media speech type