We were inspired to do this because of all of the political discourse on the internet. We wanted to create a program that would be able to point out and identify bias that is not always apparent, and wanted to do so through applications of machine learning.
What it does
Our chrome extension allows users to analyze electronic articles for possible bias and for emotionally/politically charged content. User can use the chrome extension to search for unbiased content for research and/or personal uses, allowing research to be conducted in a more educated, streamlined manner with only the click of a button. Our extension displays back to the user how biased the website is as a whole and how biased the five most common terms throughout the article are.
How we built it
First off, we built the scraper to collect and break down the content of online articles. We then integrated the two APIs, Google Cloud Platform’s Natural Language and Translation APIs, into an App Engine (also from Google Cloud Platform) to help analyze text and detect the language of the text. From there, we built the framework of the chrome extension that would be able to draw data from the article and transfer it to our App Engine, and eventually receive outputs from the App Engine to be displayed to the user.
Challenges we ran into
Accomplishments that we're proud of
We managed to finish a fully functional chrome extensions that operates exactly as desired, which is what we’re most proud of at this hackathon.
What we learned
Coming in, none of us knew anything about chrome extensions, scraping data, or App Engines. Nevertheless, we learned about each of those topics and developed our idea through integrating the different topics together.
What's next for MinusBias
- Building a website to allows users to use our technology in all forms (mobile, Safari, Firefox, etc.)
- Adding additional features to the chrome extension itself (link to redirect users to a separate website, friendlier User Interface)
- Developing the Google API to interpret text more for emotions rather than only positivity/negativity (is the author happy when discussing this topic/term? is the author disgusted?)
- Adding more languages to the interface so that users from all around the world would be able to use our chrome extension