Inspiration
We were inspired by the reality that people nowadays can tend to easily accept information they find as the truth, without delving deeper into different perspectives.
What it does
Our hack can read the HTML source on a website and use a Google API to find the top 10 results on Google when the current tab's title is searched. The extension then takes these results and outputs the titles with the links attached which can then be clicked on.
How we built it
We created a google chrome web extension utilizing JavaScript, HTML, CSS, as well as Google search API keys.
Challenges we ran into
We hoped to implement machine learning through the implementation of the fake news challenge dataset through a supervised machine learning model using a naive bayes classifier, however, this proved more difficult than we thought, and we were not able to implement it. The intention of the machine learning model was to use the top Google results and classify the text as to whether or not it "agreed", "disagreed", or "discussed" the topic of the original title.
Accomplishments that we're proud of
We all worked with new technologies that we hadn't implemented before. It was exciting to have the extension work when we pressed the button, and we were also proud of being able to use an API to automatically search for keywords.
What we learned
- How to make, edit, and debug a Google Chrome extension
- How to code in Javascript, HTML, and CSS
- How to make a Google API key
- How to connect an application to a Google API key through JavaScript code
What's next for Fact Chick
The next step would be to implement the model. From there, we would output an article that agreed, disagreed, and discussed in the extension tab. This would help encourage individuals to research arguments and topics further, and consider other perspectives.
Built With
- css
- html
- javascript
- naive-bays
- scikit-learn
Log in or sign up for Devpost to join the conversation.