The proliferation of the mobile web and the availability of large scale digital datasets has enabled a new wave of research studies that are largely driven by these new types of data generated in urban environments.
This tutorial aims to offer an overview of the opportunities and challenges posed by geolocated datasets with a particular emphasis on their use for the study of urban data science, guiding participants through the entire process of mining such datasets to using them to analyze different aspects of urban science with a social theory-backed approach. The tutorial will focus on an interdisciplinary approach to urban phenomena that integrates elements from geography, computer science, urban studies, sociology, physics and complex systems.
We will provide:
- an introduction to the basic concepts in geography and spatial statistics
- an extensive overview of some of social theories underlying the study of urban systems and how they can be employed to understand complex phenomena such as urban gentrification,
- an overview of the methodologies in the computational social science and network science to model cities
- an overview of the methodologies in machine learning as a medium to solve optimization problems and define prediction tasks in urban environments,
- a practical course on how to collect and store geo-referenced and spatial datasets,
- an overview on how to visualize spatial data and a discussion on the common pitfalls of a visual exploration of geolocated datasets.
The tutorial will be of interest to a multi-disciplinary audience that has interest in spatial data collection and analysis with an application focus in cities. Social scientists that would like to engage with geo-referenced data on a practical level, computer scientists that would like to experience a new application domain focus in urban environments and people with background in geography, architecture or planning that would like to improve their data science skills could benefit from attending. While only basic knowledge of Python is required to attend the tutorial and an understanding of the ecosystem of online services, more sophisticated programmatic examples will be described. These will be provided in an off-the-shelf manner so attendees will be able to use them even if they have an abstract understanding of the underlying systems discussed (e.g. demonstrating how APIs are used in the context of data collection won’t require knowledge of server programming).
Some familiarity with network science concepts is advisable, yet all practical examples will be covered with appropriate references and explanation of the network theory concepts. There will be limited use of machine learning algorithms through Python’s scikit-learn library to simulate a number of application examples, though sophisticated knowledge in the area won’t be required. By the end of the tutorial will have improved their programming, analytic and modeling skills in the area of urban data science. They will have experience in performing spatial data analysis using a wide variety of datasets considering different application scenarios and have example code that can be easily modified for further use.
Activities and Timeline
9:00 - 10:00
10:00 - 11:00
11:00 - 11:30
11:30 - 12:30
12:30 - 13:30
13:30 - 14:30
14:30 - 15:30
15:30 - 16:00
16:00 - 17:00
Research Session 1: Geolocated Human Behavior Through Social Media [slides]
Research Session 2: Integrating Social Theory in Human Mobility
Research Session 3: Visual Modelling of Spatial Phenomena [slides]
Practical Session 1: Analysis and Visualisation of Affordability in
Practical Session 3: Introduction to Visualising Spatial Data in Python [code]