Exposing and Explaining Fake News On-the-Fly

TL;DR

This paper presents an innovative real-time, explainable fake news detection system for social media that combines machine learning, NLP, and online lexica, validated on Twitter data with 80% accuracy.

Contribution

It introduces the first integrated system for online fake news detection that combines data stream processing, profiling, classification, and explainability.

Findings

Achieves 80% accuracy on Twitter datasets.

Provides real-time explanations of classification decisions.

Enhances trustworthiness of social media content.

Abstract

Social media platforms enable the rapid dissemination and consumption of information. However, users instantly consume such content regardless of the reliability of the shared data. Consequently, the latter crowdsourcing model is exposed to manipulation. This work contributes with an explainable and online classification method to recognize fake news in real-time. The proposed method combines both unsupervised and supervised Machine Learning approaches with online created lexica. The profiling is built using creator-, content- and context-based features using Natural Language Processing techniques. The explainable classification mechanism displays in a dashboard the features selected for classification and the prediction confidence. The performance of the proposed solution has been validated with real data sets from Twitter and the results attain 80 % accuracy and macro F-measure. This proposal is the first to jointly provide data stream processing, profiling, classification and explainability. Ultimately, the proposed early detection, isolation and explanation of fake news contribute to increase the quality and trustworthiness of social media contents.