# Automated Fake News Detection using cross-checking with reliable sources

arXiv: https://arxiv.org/abs/2201.00083 · HTML: https://mlsanity.com/paper/2201.00083/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 30 references that are in the map.
- First posted on arXiv 1 Jan 2022.

## Builds on (most cited first, 5 of 5)

- [Image-based Recommendations on Styles and Substitutes](https://mlsanity.com/paper/1506.04757/) (2015)
- [Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering](https://mlsanity.com/paper/1602.01585/) (2016)
- ["Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection](https://mlsanity.com/paper/1705.00648/) (2017)
- [Learning Word Vectors for 157 Languages](https://mlsanity.com/paper/1802.06893/) (2018)
- [Automatically Identifying Fake News in Popular Twitter Threads](https://mlsanity.com/paper/1705.01613/) (2017)

## Built on by (most cited first, 1 of 1)

- [PASTA: A Dataset for Modeling Participant States in Narratives](https://mlsanity.com/paper/2208.00329/) (2022)

## Abstract (arXiv)

Over the past decade, fake news and misinformation have turned into a major problem that has impacted different aspects of our lives, including politics and public health. Inspired by natural human behavior, we present an approach that automates the detection of fake news. Natural human behavior is to cross-check new information with reliable sources. We use Natural Language Processing (NLP) and build a machine learning (ML) model that automates the process of cross-checking new information with a set of predefined reliable sources. We implement this for Twitter and build a model that flags fake tweets. Specifically, for a given tweet, we use its text to find relevant news from reliable news agencies. We then train a Random Forest model that checks if the textual content of the tweet is aligned with the trusted news. If it is not, the tweet is classified as fake. This approach can be generally applied to any kind of information and is not limited to a specific news story or a category of information. Our implementation of this approach gives a 70% accuracy which outperforms other generic fake-news classification models. These results pave the way towards a more sensible and natural approach to fake news detection.