[Paper Review] Combating Fake News: A Survey on Identification and Mitigation Techniques

Karishma Sharma, Feng Qian+4 more|arXiv (Cornell University)|Jan 18, 2019

Misinformation and Its Impacts111 references73 citations

TL;DR

This survey comprehensively reviews fake news detection and mitigation methods, analyzes challenges, and compiles existing datasets to guide end-to-end solutions and future research.

ABSTRACT

The proliferation of fake news on social media has opened up new directions of research for timely identification and containment of fake news, and mitigation of its widespread impact on public opinion. While much of the earlier research was focused on identification of fake news based on its contents or by exploiting users' engagements with the news on social media, there has been a rising interest in proactive intervention strategies to counter the spread of misinformation and its impact on society. In this survey, we describe the modern-day problem of fake news and, in particular, highlight the technical challenges associated with it. We discuss existing methods and techniques applicable to both identification and mitigation, with a focus on the significant advances in each method and their advantages and limitations. In addition, research has often been limited by the quality of existing datasets and their specific application contexts. To alleviate this problem, we comprehensively compile and summarize characteristic features of available datasets. Furthermore, we outline new directions of research to facilitate future development of effective and interdisciplinary solutions.

Motivation & Objective

• Define fake news and characterize its dimensions and actors in information ecosystems.
• Summarize detection techniques and mitigation/intervention strategies with their advantages and limitations.
• Consolidate and describe available datasets to aid dataset selection and method evaluation.
• Discuss end-to-end system requirements and open research directions for proactive intervention.

Proposed method

• Survey and synthesis of detection and mitigation methods from the literature.
• Categorization of techniques by content features, user responses, and dissemination patterns.
• Compilation and summary of available fake news datasets and their characteristics.
• Discussion of challenges, adversarial dynamics, and practical moderation considerations.
• Proposition of end-to-end design considerations for detection and intervention workflows.

Experimental results

Research questions

• RQ1 What are the defining characteristics and dimensions of fake news across content, source, and user response attributes?
• RQ2 What techniques exist for detecting fake news and mitigating its spread, and what are their strengths and limitations?
• RQ3 What datasets are available for fake news research, and what features characterize them?
• RQ4 What are the challenges and requirements for building end-to-end fake news detection and intervention systems?
• RQ5 How can detection be balanced with moderation to ensure timely and reliable interventions?

Key findings

• Fake news can be defined to cover fabricated, misleading, impersonated, manipulated, and contextually false content with varying intents.
• Detection techniques span content analysis, social context, and propagation patterns, each with distinct limitations and performance trade-offs.
• User responses and diffusion dynamics often provide stronger signals for detection than content alone.
• There exist multiple datasets with varied annotations, highlighting the need for careful dataset selection and standardization for fair evaluation.
• End-to-end solutions require balancing aggressive moderation with rapid information access and should exploit incrementally available data to trade off timeliness and accuracy.