For our project, we wanted to connect first-gen and low income college applicants with mentors who could guide them through the college application process, and also suggest to the students a good college for them based on their interests.
What it does
We created a web-based application that takes information from both high school students and mentors. Based on the students' input, they are matched with a set of most compatible mentors using a k-d tree. Also, our program has the ability to predict a good match college using a Naive Bayes classifier. In addition, the mentors' data is collected and stored in a database for future use.
How I built it
We built our program using Python to create the backend and CSS and HTML for the front end. We process the data collected from an HTML form and processed it with a k-d tree to choose the 10 most compatible mentors for the given student. We also used a Naive Bayes classifier to train the data collected from our mentors and suggest to the student a potential college.
Challenges I ran into
None of our team members had any prior experience with HTML or CSS, so it was a challenge to figure out how to make the interface, and to communicate between the front and the back ends of our program.
Accomplishments that I'm proud of
Despite our lack of prior knowledge about webdev, our website looks fairly polished (and would look even better given more time!)
What I learned
We learned that 12 hours is not enough time to fully or satisfyingly implement an entire project!
What's next for Compass
There are a lot of directions that this project could take us in the future. Some of these include: Text parsing to allow users more freedom to describe themselves outside of checkboxes. To implement this fully, we would want to use more machine learning to train a network to match students and mentors based on a more complex algorithm. Authentication for more secure email address - security is important! Move to a cloud-based server Adaptive screen size