A Survey of Datasets for Information Diffusion Tasks

2024-07-06

arXiv.org

AI Summary

Existing research on information diffusion lacks systematic task categorization and integrated benchmark datasets. Method: Grounded in the 5W communication model, this work proposes a unified taxonomy encompassing ten subtasks—including diffusion prediction, social bot detection, and misinformation identification—and constructs the first structured, multi-task open-source dataset repository, aggregating over one hundred publicly available datasets. It further introduces a six-dimensional standardized evaluation framework covering user, network, content, credibility, temporal dynamics, and cross-platform alignment. Through comprehensive literature review, task formalization, and meta-analysis, the study identifies three pervasive data limitations: insufficient temporal completeness, poor cross-platform comparability, and inconsistent annotation practices. Contribution/Results: The paper establishes foundational infrastructure for diffusion research—namely, a principled taxonomy, a scalable dataset warehouse, and a holistic evaluation framework—while proposing actionable directions for future work, including extensible data acquisition pipelines and joint cross-dataset annotation protocols.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Application Domains: Misinformation & Fake News
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Web Mining and Content Analysis: Content-based information diffusion
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Information diffusion across various new media platforms gradually influences perceptions, decisions, and social behaviors of individual users. In communication studies, the famous Five W's of Communication model (5W Model) has displayed the process of information diffusion clearly. At present, although plenty of studies and corresponding datasets about information diffusion have emerged, a systematic categorization of tasks and an integration of datasets are still lacking. To address this gap, we survey a systematic taxonomy of information diffusion tasks and datasets based on the"5W Model"framework. We first categorize the information diffusion tasks into ten subtasks with definitions and datasets analysis, from three main tasks of information diffusion prediction, social bot detection, and misinformation detection. We also collect the publicly available dataset repository of information diffusion tasks with the available links and compare them based on six attributes affiliated to users and content: user information, social network, bot label, propagation content, propagation network, and veracity label. In addition, we discuss the limitations and future directions of current datasets and research topics to advance the future development of information diffusion. The dataset repository can be accessed at our website https://github.com/fuxiaG/Information-Diffusion-Datasets.

Problem

• Systematically categorize information diffusion tasks and datasets
• Survey datasets for prediction, bot detection, misinformation detection
• Analyze limitations and future directions in diffusion research

Innovation

• Systematic taxonomy based on 5W Model
• Categorizes tasks into ten subtasks
• Public dataset repository with six attributes