Authors: Francisco de Arriba-Pérez, Silvia García-Méndez, Fátima Leal, Benedita Malheiro, Juan Carlos Burguillo

Published on: May 03, 2024

Impact Score: 8.0

Arxiv code: Arxiv:2405.06668

Summary

• What is new: This is the first method that combines data stream processing, profiling, classification, and explainability for fake news detection.
• Why this is important: The problem of fake news rapidly spreading on social media due to the consumption of unreliable content.
• What the research proposes: An online explainable classification method that uses both unsupervised and supervised machine learning, leveraging natural language processing to profile based on creator, content, and context.
• Results: Successfully attained 80% accuracy and macro F-measure in identifying fake news on Twitter using real data sets.

Technical Details

Technological frameworks used: Online classification framework, Natural Language Processing techniques

Models used: Combination of unsupervised and supervised machine learning models

Data used: Real data sets from Twitter

Potential Impact

Social media platforms, news agencies, digital content creators, and misinformation monitoring systems.

Want to implement this idea in a business?

We have generated a startup concept here: TrustStream.