# Misleading Metadata Detection on YouTube

## TL;DR

This paper introduces UCNet, a deep learning model for detecting misleading videos on YouTube, demonstrating high accuracy and good generalization across datasets.

## Contribution

The paper presents UCNet, a novel deep network architecture specifically designed for fake video detection, and provides extensive experiments on two datasets showing its effectiveness.

## Key findings

- Achieved a macro F-score of 0.82 on FVC dataset.
- Model generalizes well across different datasets.
- Significant improvement over baseline models.

## Abstract

YouTube is the leading social media platform for sharing videos. As a result, it is plagued with misleading content that includes staged videos presented as real footages from an incident, videos with misrepresented context and videos where audio/video content is morphed. We tackle the problem of detecting such misleading videos as a supervised classification task. We develop UCNet - a deep network to detect fake videos and perform our experiments on two datasets - VAVD created by us and publicly available FVC [8]. We achieve a macro averaged F-score of 0.82 while training and testing on a 70:30 split of FVC, while the baseline model scores 0.36. We find that the proposed model generalizes well when trained on one dataset and tested on the other.