Human Mobility talk at CLISEL
Slides for my presentation at the interdisciplinary workshop on Environmental Changes and Human Mobility in Ascona.
Here's the abstract of the talk:
Understanding human mobility patterns is important in fields like migration studies, urban and transportation planning, epidemiology, and disaster response. Data on human mobility are often voluminous and difficult to interpret in tabular form; therefore, visualisation plays an important role in their analysis. One of the most widely used visualisations of such data are flow maps that represent movement between geographical locations as lines of varying thicknesses. Flow maps usually do not accurately depict the exact movement routes. Instead they are aimed at answering questions such as: Where are the origins and the destinations of the flows on the map? What are the magnitudes of the flows? Where are the largest flows? Or more complex ones like: What is the spatial structure of the flows network?
Until recently, making flow maps has involved manual drawing, or knowledge of programming or specialised software. I had learned from my experience developing flow mapping libraries that there was a need for a tool with which people without special knowledge could make flow maps and share them online. This is why I decided to develop FlowmapBlue. It is a web application for publishing interactive flow maps from data uploaded to online spreadsheets. It is free; it requires no installation and allows anyone to create a flow map in no time. Since the tool was released, people from different parts of the world have started using it to visualise various human mobility datasets and are sharing them with the general public. In the talk I will demonstrate the tool, show example datasets and discuss a few interesting patterns that can be discovered in them. I will speak about the challenges of visualisation of human mobility flows and about overcoming them. I will touch on the analysis of changes over time, on representing flows with attributes, and on the scalability of the general approach to very large datasets.