# Debunking Fake News One Feature at a Time

Source: [Debunking Fake News One Feature at a Time](<https://www.sota2.com/research/paper/68fa64a41e209131cea05d0a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:08:59.042Z

Authors: Melanie Tosik, Antonio Mallia, Kedar Gangopadhyay

Published: 2018-08-08T15:45:06.000Z

## Abstract

Identifying the stance of a news article body with respect to a certain headline is the first step to automated fake news detection. In this paper, we introduce a 2-stage ensemble model to solve the stance detection task. By using only hand-crafted features as input to a gradient boosting classifier, we are able to achieve a score of 9161.5 out of 11651.25 (78.63%) on the official Fake News Challenge (Stage 1) dataset. We identify the most useful features for detecting fake news and discuss how sampling techniques can be used to improve recall accuracy on a highly imbalanced dataset.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Fake News Stance Detection on FNC 1 (test)](<https://www.sota2.com/research/sota/fake-news-stance-detection-on-fnc-1-test.md>): Fake News Stance Detection; FNC 1 (test); Agree: 1.42.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1808.02831>)

[PDF](<https://arxiv.org/pdf/1808.02831.pdf>)