BaitBlocker

Move past the clickbait and get your information when you need it

Comment

Inspiration

Our project was inspired by the sheer amount of misinformation and clickbait that circulates around the internet in the current day. Many article titles often can be misleading and inspire readers to click onto it expecting one thing, but being met with something else. Another inspiration was how articles can also be long and drawn out, only to communicate a few simple points. As such, our project works to scrape the text in the article and revise the article to provide a summary in a way that both makes sense and covers the key points.

What it does

Our project is an unofficial Chrome extension that can be activated to extract the current URL of the page. On a valid article or text page, the client clicks a button on the extension that sends a POST request to the host device, which then sifts through the HTML and scrapes the text to collect data on the given article. The main focus of the scrape is to extract the title and the first 3500 characters to then forward it to OpenAI API with a given prompt. ChatGPT processes the data given and then generates a revised title and summary of the article.

How we built it

Most of our code was generated with the help of ChatGPT. Python, HTML, CSS, and javascript are some of the languages used to create this project. The coding was done in Visual Studio Code and then all testing was done on Google Chrome.

Challenges we ran into

Some challenges we ran into were coding errors and general errors getting the extension the work. The problems occurred both on server- and client-side. There were also countless issues with communicating between the two parties that proved to be difficult to deal with, such as issues with CORS and preflight requests.

Accomplishments that we're proud of

Some accomplishments that we're proud of include getting the project off the ground and overcoming the issues that we ran into along the way. Many of the problems we were faced with didn't include an explanation as to what was wrong, just that there was something there that wasn't working properly. Hours of research, debbugging, and troubleshooting proved to be a great feat that ultimately allowed us to complete our project and get it into a state that we can be proud of.

What we learned

One thing we learned is how much time it truly takes to get web projects off the ground and (mostly) free of bugs. As is the essence of Computer Science, only 10% of the time was devoted to the actual coding while 90% was spent troubleshooting. We also attained a better understanding of git, as well as working together as a team to divide up the work and complete each major part of the project.

What's next for BaitBlocker

In the future, if our team decides to continue working on the project, we plan on developing a Master version of our extension, called "Master BaitBlocker". It will include more features, an improved UI, and be more stable than the version that we will be presenting.

Share this project:

Updates