The following is an essay I wrote exploring different neural network architectures that can be used to detect misinformation present in social media content. The essay begins by introducing the issue of misinformation, its prevalence and the harm it inflicts in the real world. I then go on to provide a quick overview of how a basic artificial neural network functions.

Following this, I explore the academic literature on detecting misinformation across textual data, image data, multimodal sources (e.g. A video, its caption, the audio etc. - all processed simultaneously) as well as how neural networks can monitor misinformation spread.

I then end by discussing how ANNs are currently used (or underused) by social media companies and the recent approaches to managing (or not) misinformation taken by social media companies. I then conclude by suggesting ways ANNs could be implemented at greater scale to combat misinformation on social media.

This essay was submitted as an assessment for the Wider Topics in Data Science (L7) (905F3) module.