Blockchain-based rumor detection approach for COVID-19.

Area of Science:

• Computer Science
• Information Security
• Artificial Intelligence

Background:

• Social media facilitates rapid information sharing, often without veracity checks, leading to widespread false information propagation.
• Detecting rumors in massive unstructured social media data is challenging.
• Existing machine learning and deep learning methods primarily focus on detection, not prevention.

Purpose of the Study:

• To propose a novel model for the detection and prevention of transmitted rumors on social media networks.
• To leverage blockchain technology for information credibility verification.
• To develop a multi-layered framework to mitigate rumor propagation.

Main Methods:

• A four-layered framework comprising network, blockchain, machine, and device layers was designed.
• Blockchain technology was employed to verify information credibility.
• Deep learning, specifically the Bi-directional Long Short Term Memory (Bi-LSTM) model, was used for anomaly identification and rumor prevention by monitoring incoming messages.

Main Results:

• The proposed Bi-LSTM model demonstrated superior performance compared to state-of-the-art methods.
• The model achieved an accuracy of 99.63%.
• The false positive rate was significantly low at 0.13%.

Conclusions:

• The integrated framework effectively detects and prevents the propagation of rumors on social media.
• The Bi-LSTM model shows high efficacy in identifying and mitigating false information.
• This approach offers a robust solution for enhancing information integrity in online networks.