Skip to content

ChakriCherukuri/bayeshack

 
 

Repository files navigation

An Analysis of Well-Being in San Francisco

This is a one-day project completed as part of the Bayes Impact hackathon.

Prompt

Housing inequality is present in cities across the United States, rendering low income families unable to obtain affordable housing. Lack of fair housing opportunities is just one of the problems communities face: many people also lack access to transportation and services within the community. Both communities and residents suffer when specific populations cannot utilize all the resources communities have to offer. National and local data sets have been created by initiatives that are addressing the gaps between residents in communities.

Help cities enhance their use of data and evidence to uncover new ways to revitalize neighborhoods and improve the lives of residents. Leverage federal and local open data to identify disparities in access to resource, services, and housing that communities need to thrive. Interactive informative tools that show current trends, or tools to illustrate federal and local spending or regulatory changes, have the potential to reinvent the way communities come together and grow.

Original Brief for The Project

Goals

This project aims at characterizing the livability of San Francisco neighborhoods, and does so in 2 ways:

  • By providing meaningful metrics for each neighborhood, related to crime, transportation, access to restaurants, together with tools to visuzalize them
  • It investigates the relation between those features and the satisfaction of neighborhood residents.

Examples

Visualizing features of neighborhoods using chloropleths.

Chloropleth

Eploring the relationship between the features and the general level of satisfaction of inhabitants.

Screencast

Data and Methodology

The satisfaction of inhabitants is inferred from surveys and the features are inferred from census data, notably from datasf.org.

The neighborhood features include:

  • a crime index, calculated from the crime density of the area.
  • access to schools, both public and private.
  • access to restauration services, i.e. number of close restaurants and their respective ratings.
  • transportation costs, affordability, poverty, ethnicity indices taken directly from census data.

Tools

The project was developed with the Python and JavaScript programming languages. The deliverable is an interactive document provided in the form of a Jupyter notebook with advanced interactive visualizations based on leafletjs, bqplot, d3.js.

Directions for improvement

  • Incorporating historical feature and response data will allow building a neighborhood specific model that may better predict the impact of changing a feature of that neighborhood on its well-being.
  • Building a GUI using bqplot to allow the features to be adjusted visually and which displays the predicted change in well-being for the neighborhood.
  • Incorporating Yelp reviews to gauge neighborhood well-being.

Requirements

To be able to run the project the following software is required:

Geographical Data

sf_zipcodes.geojson:

A GeoJSON file that contains San Francisco zip code level topographical data. Each feature contains an attitude id which is the zip code associated with the Polygon.

About

Bayes hack 2016, the quants

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 99.7%
  • Python 0.3%