- Hackathons
- Builders
- Organizations
- Learn
Description
Links
- https://github.com/viviansongg/bias.io
Inspiration
The inspiration behind developing an application that uses AI to analyze the credibility of online articles stems from the growing challenge of misinformation in the digital age. With the proliferation of content on the internet, distinguishing between reliable information and false or misleading narratives has become increasingly difficult. This application leverages advanced AI algorithms to assess various credibility factors, including the source's reputation, the article's language and tone, and the presence of supporting evidence. By providing users with credibility scores and detailed analysis, bias.io aims to empower individuals to make informed decisions and foster a more trustworthy online information ecosystem.
What it Does
This program evaluates articles' credibility based on various criteria including key phrases found in the articles, the political reputation of the site, source credibility, author/publisher reliability, site extension and more. Through a dynamic and intuitive user interface, bias.io aims to promote media literacy and critical thinking, and combat misinformation in the digital age.
How We Built It
bias.io was developed using React and Flask (HTML/CSS/JavaScript) to build the web application framework using components from Chakra UI, as well as Python and Flask for the back-end. Additionally, it leverages the Gemini API to generate bias-free article summaries and relies on News API to provide a database of articles to operate on. Finally, the platform was deployed using Github Pages.
Challenges We Ran Into
While much of the backend was relatively seamless, we ran into some difficulties training Gemini using existing articles. We also had some challenges with implementing the front-end as some of our team members ran into issues hosting the application on the web to test various GUI components.
Accomplishments That We're Proud Of
We are proud to have completed a functioning prototype that successfully integrates Gemini's AI model to perform a detailed analysis on article credibility.
What We Learned
This project allowed us to learn how to use the Gemini API, News API and Flask.
What's Next?
Upon expanding on the project, we would use NLP to train the AI model on a diverse set of articles to develop more efficient algorithms for determining credibility.