Fake News Detection

Homepage of our website
Scrolling down on the homepage to see what the website offers
The validate news tab that allows user to input their content and validate it
A trusted sources guide that links users to different reputable sources

Our Team

High school students who are studying computer science. We hope to build our experience and knowledge in coding with this platform we created.

Inspiration

As information becomes widespread, deciding whether the information is factually true can be challenging. Students like us who rely on the internet and are exposed to thousands of pieces of information, including images, articles, and other types of information, every day. Because of this, knowing how trustworthy the source or the facts that we are looking at is important. This is why we built our fake news detection that allows users to verify their news.

What it does

Users can paste a news article, a claim, or a webpage URL directly into the input field. The application sends this content to the Google Gemini AI API, which analyzes it and returns: A truthfulness score from 1-100, accompanied by a label such as “Likely True”, “Misleading”, or “Likely False” A written explanation that lists specific reasons behind the grading A trusted sources guide built into the app, listing reputable news outlets and Wikipedia, so users can cross-check information on their own The goal of this project is to give users a quick, AI-assisted check on suspicious content on the internet

How we built it

We first decided how we were going to verify the news. We ultimately decided to use APIs. However, none of us had experience with APIs, forcing us to learn from scratch. As we learn, we also design the interface and build it alongside learning how to use APIs. Once we understood how the API works, we then incorporated it into our website.

Challenges we ran into

As high school students, one of our biggest challenges was learning Python web development while building a real-world application. None of us had extensive experience with FastAPI, server-side development, or integrating AI APIs, so we had to learn these concepts from the basics. Another challenge was designing prompts that would encourage the Gemini API to return consistent, well-structured analyses instead of vague responses.

Accomplishments that we're proud of

Our accomplishments that we’re proud of are that we’re able to learn and use an API from scratch. At the beginning, none of us had any knowledge about APIs. We were researching what to use on how to verify our information when we stumbled upon APIs, which is when we decided to use them. In the end, we went from not knowing what it was to being able to understand and incorporate it in a working project.

What we learned

We were able to learn how API works and incorporate them in a working project. Besides API, we learned how to work as a team, but also learned how to use outside resources as a tool to help, rather than a place for plagiarism.

What's next for Fake News Detection

Fake News detection is our first project incorporating APIs. Our goal in the future is to become familiar with tools like these, but also to learn new concepts. Our goal is to use what we learn to provide useful resources like these in the future.