We were inspired by the notion that correlation does not imply causation, and we set out to test the truth behind this statement. After spending 2 of our 1 hour brainstorming budget throwing ideas at each other, we decided on combining star alignment with machine learning. The outcome? This application. Pleiades. Named after The Seven Sisters star system, part of the constellation Taurus. In this day and age, many people face the problem of finding a true, deep connection with others. What could be deeper than the connection we have with the Universe?
What it does
This is a dating application. There are many like it, but this one is ours, and it contains several key differences. Our dating app utilizes dark maths and powerful star tracking to calculate everyone's unique Astral Love Quotient. Separate from other dating apps that rely on decades old static data, ours stays up-to-date by constantly recalculating based on the movement of the Pleiades star cluster. In addition, our highly classified ALQ algorithm takes into account various cosmic factors such as the frequency of the universe to ensure that any factor that affects your love life is accounted for.
How we built it
Our application relies on Tangram for easily training our machine learning algorithm. To train our model we used data from the General Social Survey, which is collected by the National Opinion Research Center at the University of Chicago. By selecting key parameters and doing thorough data cleaning using the software package Stata, we were able to confidently feed our model thousands of data points and guarantee a match. Using Flask we were able to connect our python backend to a HTML, CSS, and JS frontend.
Challenges we ran into
The stars are fickle things, and so is code. The main challenges we faced were communicating with our website through python and figuring out how we could send and receive data from users that sign up for our astral love service. We burned through many different ideas for how to do this since none of us have any experience with web development beyond basic HTML and CSS. There were unfortunately a few moments where we spent precious hours on a single problem that had a retrospectively easy solution.
Accomplishments that we're proud of
We are proud of our ability to combine seemingly unrelated aspects of the universe as well as seemingly unrelated programs. This is our first hackathon and we are excited to share what we have been able to put together, and the coding challenges we surmounted.
What we learned
We have learned that brainstorming is hard, but when there is an idea that motivates everyone on the team, people will work with passion to meet goals. Also, web development is a pain on such tight time constraints when no one has much in-depth experience with it.
What's next for Pleiades
Pleiades is a wonderful concept, although it has its flaws. The GSS dataset, while super cool, is probably not the best for determining someone's love life, and we did not have ample time to vet it properly. Also our application lacks many features we had planned but couldn't achieve, such as a proper database for ~real~ users and not just accounts we created ourselves.