Skip to content

bmbejcek/TrafficCruising-DSSG2017

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

51 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Traffic Cruising Data Science for Social Good Project

Description

Vehicle cruising (individuals looking for parking and for-hire vehicles operating without a passenger) is a major contributor to traffic congestion in downtown Seattle. Still, the magnitude and location of vehicle cruising is poorly understood. To get a better understanding of where vehicles cruise, we propose a framework for using traffic sensor data. We generate most likely paths traversed through filtering out unrealistic behavior and incorporating routing. We break up individual trips via segmentation in terms of time and method of transportation. After segmentation, we introduce metadata to describe the trip and use a semi-supervised machine learning approach to label the data. Ultimately, we create an interactive heat map of downtown Seattle that can be used to visualize the relative levels of cruising.

This research has the potential to help transportation agencies, technology companies, and car companies predict the availability of parking and more accurately direct travelers with online, mobile, and connected tools, thereby reducing congestion impacts, emissions, and fuel costs.

Web Application Demo

Description of Folders

  • analysis: contains code for supporting tasks carried out throughout the process of building the pipeline.
  • app: contains code for the web application to visualize the aggregated data.
  • data: contains supporting data necessary at different steps in pipeline.
  • models: contains analysis for different machine learning approaches.
  • pipeline: contains code for our process of transforming the data to a usable format.
  • results: contains final papers, presentations, and images.

How To Use Web Application Demo

  1. Clone github repository to local computer
  2. Download Python 3.6
  3. Install required dependencies with the following command: "pip install Flask"
  4. Navigate to app folder
  5. From command line, type: "python app.py"
  6. Open web browser to "localhost:5000"

Team Members

DSSG Fellows

  • Brett Bejcek, The Ohio State University
  • Anamol Pundle, University of Washington
  • Orysya Stus, University of California, San Diego
  • Michael Vlah, University of Washington

Data Science Leads

  • Valentina Staneva, eScience Institute
  • Vaughn Iverson, eScience Institute

Project Lead

  • Steve Barham, Seattle Department of Transportation

About

Public Repository for Traffic Cruising DSSG 2017 Project

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 52.4%
  • JavaScript 16.7%
  • R 15.9%
  • Jupyter Notebook 12.1%
  • HTML 1.6%
  • CSS 1.3%