TSA 2016 - Summer School on Topological and Scaling Analyses of Transport and Social Media

Summer School on Topological and Scaling Analyses of Transport and Social Media
Data (June 13?17, 2016, Gävle, Sweden)
Current geospatial analysis is highly constrained by Euclidean geometry and Gaussian
statistics in the sense of geographic locations and distances, as well as a well-defined mean
and small variance. These two ways of thinking (Euclidean geometry and Gaussian statistics)
suffer from some major disadvantages that prevent us from gaining new insights into the
underlying complexity of geographic phenomena. On the other hand, topology, fractal
geometry, and power law statistics represent new perspectives for geospatial analysis,
particularly in the era of big data.
This summer school, sponsored by COST Action TU1305: Social networks and travel
behavior, and by the University of Gävle, will provide a set of workshops on alternative
geospatial analysis methods surrounding topological and scaling analyses that will enable us
to better understand structure and dynamics of transport systems and human mobility
patterns. The methods include head/tail breaks, topological analysis, complex networks, and
agent-based simulations. The summer school is further motivated by the emerging big data
harvested from the Internet and social media such as OpenStreetMap, Flickr, Twitter, and
Brightkite, which provide a new instrument for transport research and human mobility. Big
data differs fundamentally from small data in terms of basic data characteristics such as
accurately measured or roughly estimated, individual-based or aggregated, and massive or
small amounts. Big data also differs fundamentally from small data in terms of the data
analytics that surround these two ways of thinking geometrically and statistically.
The summer school will combine lectures, hands-on exercises, and projects. We will provide
related data obtained from OpenStreetMap, Twitter, Brightkite, and Gowalla for the hands-on
parts. We will offer brainstorming sessions to develop research projects that could hopefully
lead to joint publications among the participants. The hands-on exercises will be based on the
tools listed below, but the summer school is focused on research rather than software training.