LIAR is a publicly available dataset for fake news detection. A decade-long of 12.8K manually labeled short statements were collected in various contexts ...

LIAR is a publicly available dataset for fake news detection. A decade-long of 12.8K manually labeled short statements were collected in various contexts from ...

LIAR is a dataset for fake news detection with 12.8K human labeled short statements from politifact.com's API, and each statement is evaluated by a politifact.

LIAR Dataset, is a new fake news detection dataset that includes 12.8 thousand short phrases labeled by hand for honesty, topic, context/place, speaker, ...

Kaggle is the world's largest data science community with powerful tools and resources to help you achieve your data science goals.

In this paper, we present LIAR: a new, publicly available dataset for fake news detection. We collected a decade-long, 12.8K manually labeled short statements ...

A new benchmark dataset for fake news detection, to appear in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017 ...

A coarse-to-fine cascaded evidence-distillation neural network for explainable fake news detection.

The LIAR Dataset Dataset contains 12.8K manually labeled short statements in various contexts from POLITIFACT.COM, which provides detailed analysis report.

LIAR-PLUS is a benchmark dataset for fake news detection, released recently. This dataset has evidence sentences extracted automatically from the full-text ...