An End-to-End Framework to Identify Pathogenic Social Media Accounts on\n Twitter

2019/05/04 by Elham Shaabani, Ashkan Sadeghi-Mobarakeh, Shaabani, Elham +5
Computer Science · Social Sciences · #FOS: Computer and information sciences #Hate Speech and Cyberbullying Detection #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Misinformation and Its Impacts #Social and Information Networks (cs.SI) #Spam and Phishing Detection

paper · pdf · doi:10.48550/arxiv.1905.01553

openalex publication_date 2019/05/04 · openalex created_date 2022/07/23 · openalex updated_date 2026/07/28

Abstract

Pathogenic Social Media (PSM) accounts such as terrorist supporter accounts\nand fake news writers have the capability of spreading disinformation to viral\nproportions. Early detection of PSM accounts is crucial as they are likely to\nbe key users to make malicious information "viral". In this paper, we adopt the\ncausal inference framework along with graph-based metrics in order to\ndistinguish PSMs from normal users within a short time of their activities. We\npropose both supervised and semi-supervised approaches without taking the\nnetwork information and content into account. Results on a real-world dataset\nfrom Twitter accentuates the advantage of our proposed frameworks. We show our\napproach achieves 0.28 improvement in F1 score over existing approaches with\nthe precision of 0.90 and F1 score of 0.63.\n

Related

• Early Identification of Pathogenic Social Media Accounts
• Causal Inference for Early Detection of Pathogenic Social Media Accounts
• Less is More: Semi-Supervised Causal Inference for Detecting Pathogenic Users in Social Media
• Detecting Pathogenic Social Media Accounts without Content or Network Structure
• A Feature-Driven Approach for Identifying Pathogenic Social Media Accounts