NeighborhoodPoweredByPreference

• 

  House Search

• 

  Interactive Map

Inspiration

We know that purchasing a home is a big investment for any individual, our goal for this project was to give the average person the chance to choose the perfect home based the on their neighborhood preferences. The inspiration of this project came from building an application that allowed users to specify what is most important to them which does not currently exist. We believed that big data plays an important role in today’s world and it should be utilized to provide meaningful information.

What it does

The user can select preferences for their ideal neighborhood and the application will generate a gradient that represents the cities that meets the user’s most important preference. The preferences that can be selected includes median resident age, median household income, estimated per capita income, the mean price of all housing units, the mean price of townhouse, the monthly rent, cost of living index, population density, average property tax with mortgages, unemployment rate and the resident to sex offender ratio.

How we built it

We used Jupyter Notebook and python to clean, configure and analyze the data. The individual city data sets for each preference were collected from the city-data.com website and we retrieved the longitude and latitude for each cities’ area in a geoJson format to provide the visualization on an interactive map.

Challenges we ran into

We spent hours brainstorming on an idea. Our original plan was to collect data from cities across the united states but with the amount of data that needed to be gathered, we knew with the time limitation, this was impossible to do. Therefore, we decided do it on a sample scale focused on the Maryland cities. We also did not have much experience with Python and Jupyter Notebook to create the application. The other challenge was working in a small group of two people, who did not have all the skills to develop this application therefore we had to learn quickly and apply what we learned.

Accomplishments that we're proud of

We were able to work on a project that was of interest for both of us. It utilized machine learning specifically decision trees and data analytics to provide a useful resource for people deciding on a neighborhood to purchase a home in. After working with no sleep and painstaking hours, we realize the lack of data and structured data was a challenge and even with that, we did our best with the data we had.

What we learned

We learned to gather data and make data meaningful and easily accessible (visually appealing) for people. We learned to use new tools such as Jupyter Notebook with Python, geoJson format and built in libraries. We learned to work as a team even under pressure and was able to utilize the resources provided to use such as mentors and sponsors.

What's next for NeighborhoodPoweredByPreference

Despite the time restraint, we still want to continue to build a larger scale application that will allow the users to have access to information for cities/neighborhoods around the country (and maybe the world). We wanted to add more preferences to the data set to make the information provided to users even more accurate based on their personal preferences. Also, continue building a stronger machine learning model.

Built With

• city-data
• excel
• json
• jupyter
• python