GEOCYCLE@WWW 2017

April 4th, Perth, Australia

What IS geocycle about?

Geography plays an important role in everyday life, and many of the decisions people take depend on where they live, where they are now, and the locations they are familiar with. The main objective of the GEOCYCLE tutorial is to arm attendees with both theoretical and practical knowledge about the whole process of making sense of geospatial data, rather than focusing on specific technologies, tools or data sources. Our aim is to provide a broad vision of all the processes and technologies available to researchers, and to center on the basics of geospatial understanding to convey best practices when dealing with such data. We present how geospatial data is different from other kinds of data, and therefore requires special considerations when representing, processing, modeling, analyzing and visualizing.

MOdules

1. Perception

We present an introduction into modern geography theory, where we dive into specific areas of the geography literature that are particularly relevant to the WWW audience. We further describe how machines commonly represent geographic data and how humans in turn perceive this data. This includes discussions on how people discuss and create places beyond simple coordinate reference systems.

2. Analytics

We first cover basic techniques for operating on geotagged data, such as determining the distance between geographic coordinates, and computing areas of and overlaps between polygons. We then move to more advanced techniques, such as clustering and density estimation. We particularly show how geotagged data differs from traditional data and thus often requires special considerations in order to obtain reliable output, such as understanding which statistical techniques are (not) appropriate.

3. Visualize

We cover a variety of techniques the attendees can use to visualize and explore actionable insights from geotagged data. A hands-on session will let the attendees first interact with real geotagged data to get familiar with visualizing a number of data representations using projections, and then present several use cases for them to investigate using suitable techniques.

Where and when

Tuesday April 4th 2017

Perth Convention and Exhibition Centre (PCEC)

Vocus Executive Boardroom (floor plan)

Program

9:00-10:00 - Perception (David Shamma)

10:00-10:30 - Analytics (Bart Thomee)

10:30-11:00 - Coffee break

11:00-11:30 - Analytics (Bart Thomee)

11:30-12:30 - Visualize (Rossano Schifanella)

12:30-13:30 - Lunch break

13:30-17:00 - Practical sessions (part 1)

15:00-15:30 - Coffee break

15:30-17:00 - Practical sessions (part 2)

Material

Slides:

1-perception.pdf
2-analytics.pdf
3-visualize.pdf


Speakers

Rossano Schifanella

University of Turin

Bart Thoomee

Google

David Shamma

Centrum Wiskunde & Informatica