Modern search engines and browsers excel at finding content that suites users preferences. However, as consequence of their precision users are becoming more unconsciously biased in terms of the media they consume. To counter this, we created Elder^_^Bro (the counter-opposite of Big Brother) to help users reflect on the content they are (and aren't) exposing themselves to in order to promote a more open-minded online community.
What it does
By activating a session with Elder^_^Bro, he will accompany you as search through various websites, and he will note down the biases of the webpages you view. Afterwards, he'll give you a report about the bias of your overall session.
How we built it
Challenges we ran into
At first we struggled with breaking down the problem into components that the NLP API could manage. Another issue we ran into, we determining how we should receive the input of the website URLs. In order to allow a seamless user experience, we decided to use a chrome extension to allow Elder Bro to work in the background.
Accomplishments that we're proud of
We're proud of learning how to use Natural Language Processing with Cohere as none of us had experience with Machine Learning before. We got a better intuition of how to use ML to solve problems.
What we learned
The most unfamiliar aspect of this project was Cohere's NLP API. We had to do a lot of research of how to configure the provided models in order to suit our needs. Many members of our team also explored Chrome extension design for the first time to implement the feature of browsing users' histories.
What's next for Elder^_^Bro
In the future, we'd like to expand the library of categories that Elder Bro can keep track of and allow users to reflect on more biases in their everyday content consumption. We'd also like to gamify the website, by having Elder Bro reward users with badges of honor for maintaining a balanced browsing history.