Spanish Fake News Corpus With Paired True And False Articles

The Spanish Fake News Corpus Version 2.0 contains pairs of fake and true publications about different events, all written in Spanish. It was collected from November 2020 to March 2021 from various web sources, primarily newspapers and media.

Use Cases

• Train a binary classifier to distinguish between fake and true publications using the paired article structure.
• Analyze linguistic patterns in Spanish fake publications versus true publications from the specified time range.
• Build a model to detect fake news events by leveraging the event-based pairing of articles in the corpus.

Strengths

• Contains pairs of fake and true publications for direct comparison.
• Data collection occurred over a defined 5-month period from November 2020 to March 2021.
• All content is in Spanish, providing a resource for a specific language.

Limitations

• The dataset size, row count, and column structure are unknown, limiting reproducibility.
• Geographic or source bias is likely as data was gathered primarily from web-based newspapers and media.
• The temporal coverage is limited to a 5-month window, which may not capture long-term trends.

Provenance

Source

sayalaruano on Hugging Face.

Collection Method

Collected from various web sources, primarily newspapers and media.

Time Range

November 2020 to March 2021.

Freshness

Last updated on 2022-03-22.

Geography

null