iflament / datascienceforsocialgood Goto Github PK

DSSG Fellows: Io Flament, Momim Malik, Cristina Lozano
Technical Mentor: Qiwei Han
Project Manager: Laura Szczuczak

This project was conducted as part of Data Science for Social Good (DSSG) Europe 2017 fellowship, further details of the twelve week summer fellowship can be found here: https://dssg.uchicago.edu/europe/

Project website with interactive visualizations: http://dssg-eu.org/florence

Mass tourism - a form of tourism that involves thousands of people visiting the same area - is on the rise. High-speed and low-cost means of travel allow larger amounts of people to travel faster and more frequently than ever before. Online resources, social media, and mapping applications are making it easier to choose a destination and plan an itinerary.

While tourism is an economic asset to communities worldwide, many touristic destinations are ill equipped to respond to the increasing flux of visitors. The unprecedented number of tourists is causing concern among local, regional and national goverment agencies. Cities working with analog management of their cultural resources are searching for sustainable solutions that will benefit both tourists and residents alike.

During summer 2017, our team at Data Science for Social Good analysed the spatial and temporal patterns of tourist movements within the city of Florence (data from summer 2016), one of Europe's oldest and most beautiful historical landmarks. This attempt to quantify and describe the extent of mass tourism in the region is one of the several ongoing intiatives that local government and tourism agencies are taking to better manage the influx of thousands of visitors, improve decision-making and maintain public safety.

This repo contains the code developed to run the spatial and temporal analyses of the project, and generate informative dynamic visualizations, on multiple civic data sources. Questions we sought out to answer included: What locations in Florence are the most crowded, and at what times (hours of the day / days of the week / dates over the summer)? Where do people transition to and from? How long do they stay in each place? What are measures & policies that could be introduced to "reroute" some of the crowds to avoid mass aggregation at the same areas & times?

We created interactive visualizations using Uber's DECK-GL library aggregate the movements of users from the different data sources, in time and space, and create 3 dimensional representations of crowding in the city.

Full page here: http://dssg-eu.org/florence/fountain.html