Inspiration
We were inspired by our Machine Learning class where we have analyzed other, though far-less robust, datasets previously. Especially given the current pandemic, housing costs can be a tremendous burden for many individuals who are struggling or are in poverty. Learning more from this dataset could help government gain useful information about where people need the most help.
What it does
We have used a dataset of U.S. housing prices to analyze and visualize the data to learn more from it, including correlations between prices and other factors that correspond to accessibility, such as wheelchair access. Our goal was to also then train a model to predict housing prices using all of the useful features (we scrapped some of the url features), but we unfortunately ran out of time.
How we built it
We built it using python and sklearn python package
Challenges we ran into
This was a large dataset, which means there is potentially more useful information, but there's also a lot of non-useful information to sort through. Additionally, larger datasets just take longer to work with.
Accomplishments that we're proud of
We feel that these results (or ones from similar projects) could really make an impact if governments are able to use the data effectively to help those who need it the most
What we learned
We learned a lot about the housing market in the U.S. during this hackathon, as well as gained more experience training ML models
What's next for We want the 2%
More ML projects!
Built With
- python
- sklearn
Log in or sign up for Devpost to join the conversation.