πŸ’€πŸ™A map of death in New York City
Jupyter Notebook JavaScript CSS HTML
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
dist
img
src
.babelrc
.gitignore
package-lock.json
package.json
readme.md
webpack.config.js
webpack.config.production.js

readme.md

Death Map NYC

This is the source code for deathmapnyc.com, a map of deaths that occurred in New York City in 2015 and 2016 based on available data. For some categories/data-sources there's data going as far back as 2006. Check the "Show All Years" button to see that.

Data Sources

You can see how the data was combined and used in the jupyter notebook.

If you have suggestions of other data sources, please let me know by opening an issue.

Notes

Rather than use clustering to show where lots of death occurred at the same location, this map shows each record but adjusts the locations slightly so they're visible.

Development

To get started:

# clone it
git clone git@github.com:zischwartz/deathmapnyc.git && cd deathmapnyc
# install deps
npm install
# start development webpack dev server
npm start
#
# open http://localhost:3333/

If you have docker (docker for mac) installed , you can easily run the jupyter notebook that collects and processes the data.

npm run jupyter
# follow instructions, copy and paste link into browser, and then click process.ipynb

Limitations & Caveats

  • The data is incomplete, there are many ways people die that aren't captured by any of the four data sources that were available.
  • Some records contain multiple deaths, especially the OSHA and motor vehicle records. These still appear as one icon on the map.
  • The data is messy, some deaths aren't displayed because there isn't enough information.
  • The location data is particularly messy, and is often the result of geocoding an address in the jupyter notebook using google maps, which may or may not be accurate.
  • OSHA records may in some cases be the address of the employer rather than the location of the accident.

2015 Version

This is my second attempt at this, the first version from 2015 is here(source code) but it may break your browser.