Skip to content

Hack Arizona 2019 Submission, a webapp that predicts safe routes for pedestrians using machine learning and data-driven decision making.

License

Notifications You must be signed in to change notification settings

mikesinouye/safeStroll

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

safeStroll

Hack Arizona 2019 Submission, a webapp that predicts safe routes for pedestrians using data-driven decision making. Our crime data comes from the Tucson Police Department and the University of Arizona Police Department incident datasets. Both were accessed through Socrata. We parsed this data for violent crimes such as assaults, homicide, and robberies. From here, we extracted the latitudes and longitudes of the crime location and assigned weights to the crime type. Homicides were weighted the highest, assaults were weighted lower based on other attributes like whether a deadly weapon was used or it was a sexual assault.

TPD incident dataset: https://moto.data.socrata.com/dataset/Tucson-Police-Department/5e96-55x5

UAPD incident dataset: https://moto.data.socrata.com/dataset/The-University-of-Arizona-Police-Department/tg4d-cwgz

Streetlights dataset: http://gisdata.tucsonaz.gov/datasets/streetlights-city-of-tucson-open-data

Bicycle traffic incident dataset: https://azbikelaw.org/crashmap-data/

Instructions for set-up to run yourself:

  • Runs on node.js
  • Install the appropriate packages in server.js to get it to run properly
  • Once that is done you can start that by typing in node server while in the backend
  • Use port 9190

Instructions for Android:

  • Move apk onto android device using flash drive
  • Install by clicking on the download
  • From there it should take you to website

About

Hack Arizona 2019 Submission, a webapp that predicts safe routes for pedestrians using machine learning and data-driven decision making.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 99.3%
  • Other 0.7%