Inspiration

We were inspired by that webpage lauren was talking about

What it does

We built a chrome extension that asks the user if they would like to scan the page for biased content. If the user chooses "Yes", the extension takes in the link of the current page and runs it through a web scrapper we made with Aylien Text Analysis to operate all the paragraph text on the web page from other content in the page. We then use several API's to read through the content, taking into account subjective words, emotionally charged punctuation, instances of possible sarcasm, title analysis, and negative/positive emotion. After obtaining that information, we take the ratio of subjective words to objective words and score it on bias, in order to give the user an educated expectation of what they will be reading in that specific article. The charged language will also be highlighted on the web page to give the user a more visual comparison of the score displayed by the plug in and what they will be reading. **lists

How we built it

We used Javascript, HTML, CSS, and JSON to make the Chrome extensions and Node.js, Python, and Elastic Search to build the backend logic that outputs bias results**.

Challenges we ran into

We were deeply challenged by both MongoDB and Elastic Search. Getting either database application running and connecting them with our own functions was challenging but very educational. We also had some difficulty debugging and opening a new pop up extension and wifi connection.

Accomplishments that we're proud of

Making a chrome extension, understanding more about Javascript/Node.js, and creating an API

What we learned

We learned that databases can be very challenging to work with, but can really help make our program more efficient. We also we learned to code some JavaScript and using it with Node.js.

What's next for Bias Flagger

We hope to get the extension published online as well as well as create a more detailed documentation on what makes the analyzed document more biased than others so many people will be able to not be falsely influenced by one sided news articles and political propaganda.

Share this project:

Updates