Skip to main content

YouCred: Combatting Fake News with AI

This project is a tool for Pesa-Check to find and mitigate the spread of fake news using Machine Learning and Natural Language Processing.

View Project Demo
  • User Research
  • Machine Learning
  • Fullstack Development
Pesa Check's result page and analytics graph.

Mitigating Fake News with Machine Learning

PesaCheck is a fact-checking organization in Kenya that aims to find and mitigate the spread of fake news. We partnered with them to create a tool that would help them find fake news and analyze the spread of fake news on different social media platforms like Youtube and Google Search.

We have created a tool that uses Machine Learning and Natural Language Processing to find fake news and analyze the spread of fake news using audio and text data.

The layers sidebar design, now with user profiles.
Multiple user annotations on a shared layer.

Creating an Advanced Search Engine

A problem that many fact-checking organizations face is that they have to search between different search engines and platforms. The tool that we have created allows users to search between different platforms like Youtube and Google Search using tags and labels on the content.

Upload annotation screen.

Scalable Tags & Labels

Marking and annotating areas on high resolution biomedical images is the core experience of the app, and it was easy to get lost or lose sense of scale when zooming in on details. Adding measurements for the perimeter and area of an annotation both helped to communicate the overall scale of the image and how large the annotated feature is in comparison.

Project outcomes

We worked with election officials to test the tool and received positive feedback when working with the 2021 Kenya elections. We were able to track the amount of fake news circulating difficult topics like Ebola and were able to mitigate the spread of fake news.