Detecting Fake News on Social Media: A Novel Reliability Aware Machine-Crowd Hybrid Intelligence-Based Method

TL;DR

This paper introduces a novel reliability-aware hybrid intelligence method for fake news detection on social media, integrating machine and crowd intelligence with reliability modeling to improve detection accuracy.

Contribution

The paper proposes the RAHI method that uniquely incorporates reliability assessment into hybrid fake news detection, enhancing performance over existing methods.

Findings

Demonstrates effectiveness on Weibo dataset

Outperforms existing fake news detection methods

Provides reliable predictions with confidence measures

Abstract

Fake news on social media platforms poses a significant threat to societal systems, underscoring the urgent need for advanced detection methods. The existing detection methods can be divided into machine intelligence-based, crowd intelligence-based, and hybrid intelligence-based methods. Among them, hybrid intelligence-based methods achieve the best performance but fail to consider the reliability issue in detection. In light of this, we propose a novel Reliability Aware Hybrid Intelligence (RAHI) method for fake news detection. Our method comprises three integral modules. The first module employs a Bayesian deep learning model to capture the inherent reliability within machine intelligence. The second module uses an Item Response Theory (IRT)-based user response aggregation to account for the reliability in crowd intelligence. The third module introduces a new distribution fusion mechanism, which takes the distributions derived from both machine and crowd intelligence as input, and outputs a fused distribution that provides predictions along with the associated reliability. The experiments on the Weibo dataset demonstrate the advantages of our method. This study contributes to the research field with a novel RAHI-based method, and the code is shared at https://github.com/Kangwei-g/RAHI. This study has practical implications for three key stakeholders: internet users, online platform managers, and the government.