How Do Social Bots Participate in Misinformation Spread? A Comprehensive Dataset and Analysis

2024-08-18

arXiv.org

AI Summary

Social bots amplify misinformation dissemination on social media, yet their interaction patterns with misinformation remain poorly understood—particularly on Chinese platforms like Weibo. Method: We propose a weakly supervised framework for automatic bot labeling and construct the first large-scale, multimodal dataset integrating high-accuracy bot detection (65,749 bots) with fine-grained misinformation annotation (11,393 fake vs. 16,416 real posts). Our approach fuses multimodal features, models user behavioral dynamics, and jointly analyzes bot–misinformation interactions. Contribution/Results: We systematically uncover how bots intensify echo chambers and generate manipulative content at scale. Experiments demonstrate high labeling accuracy and reveal that bots propagate misinformation significantly more aggressively than humans. This work provides both a novel methodological framework and critical empirical evidence for combating online misinformation.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Application Domains: Misinformation & Fake News
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Mitigating misinformation and disinformation
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

The social media platform is an ideal medium to spread misinformation, where social bots might accelerate the spread. This paper is the first to explore the interplay between social bots and misinformation on the Sina Weibo platform. We construct a large-scale dataset that contains annotations of misinformation and social bots. From the misinformation perspective, this dataset is multimodal, containing 11,393 pieces of misinformation and 16,416 pieces of real information. From the social bot perspective, this dataset contains 65,749 social bots and 345,886 genuine accounts, where we propose a weak-supervised annotator to annotate automatically. Extensive experiments prove that the dataset is the most comprehensive, misinformation and real information are distinguishable, and social bots have high annotation quality. Further analysis illustrates that: (i) social bots are deeply involved in information spread; (ii) misinformation with the same topics has similar content, providing the basis of echo chambers, and social bots amplify this phenomenon; and (iii) social bots generate similar content aiming to manipulate public opinions.

Problem

• How social bots accelerate misinformation spread on Sina Weibo
• Constructing a multimodal dataset to analyze bot and misinformation interplay
• Examining bot roles in echo chambers and opinion manipulation

Innovation

• Constructed large-scale multimodal misinformation dataset
• Proposed weak-supervised annotator for bot detection
• Analyzed bot-amplified echo chambers in misinformation