BoilerMatch

New Year, New You! Find that special someone for U!

Comment

Inspiration

As freshman college students, we found ourselves and our fellow BoilerMakers sometimes struggling to meet new people and make new friends as well as engage in the Purdue Social Scene. And with Valentines Day coming up, we decided to base our project theme on it.

What it does

It creates a mathematical model of a user using a questionnaire which is then used to match users with other Boilermakers by computing likelihood compatibility both in terms of personality and ideology.

How we built it

We utilized HTML/CSS with Javascript to create a website that users would fill out a questionnaire on. This data was then sent to a Google Cloud Database and then analyzed using Python and Pandas to compute the most compatible matches. The matches are then returned and shown to the user using Python and Flask in conjunction with HTML.

Challenges we ran into

We had to first develop a unique questionnaire that could capture a good sense of a person's personality and evaluate it quantitively. After this we had to develop a unique mathematical process to evaluate the user data and generate the compatibility with other users.

Accomplishments that we're proud of

We managed to develop a full, working pipeline for our project that allowed us to test and implement technologies and frameworks we have never used before in conjunction with multiple programming languages. We also developed a mathematical function to display the combability scores in a way that could be better understood by humans.

What we learned

We learned how to use HTML/CSS with JavaScript to create a website and send data to a Google Cloud Database. We then learned how to read and manipulate this data using Python and Pandas and then returned to the user to be displayed using a mix of Python, Flask, and HTML/CSS.

We also learned about the considerations that go into not only figuring out the type of database and dataset we wanted to work with, but how to create a way to get quantitative data to represent categorical information and then a subsequent method to evaluate said data.

What's next for BoilerMatch

We plan to move it from a localhost to be purely run on the cloud so any BoilerMaker, anywhere at anytime, can meet and fine new people. We also plan to overhaul the UI and provide users with greater customizability to allow for matches with increased compatibility.

Share this project:

Updates