This course will introduce you to the nascent field of Geographic Data Science using the industry standard Python programming language.
We will cover the key steps involved in solving practical problems with spatial data: design, manipulation, exploration, and modelling. These topics will be explored from a “hands-on” perspective using a modern Python stack (e.g. geopandas, seaborn, scikit-learn, PySAL), and examples from real-world spatial and tabular data.
We will start with an overview of the main ways to access and read spatial data formats such as shapefiles or GeoJSON from disparate sources. Then we will move on to techniques to visualise (e.g. choropleths) and summarise your data, including exploratory spatial data analysis techniques. From there we will cover traditional as well as explicitly spatial unsupervised learning (clustering). The course is intended to provide practical support to researchers and practitioners by introducing them to useful strategies to learn more from their spatial data. There will be time for self-directed learning using data from the CDRC data store.
- Basic understanding of Python as a programming language for data science
- Introductory-level use of the workhorse environment (i.e. Jupyter) and libraries in Python for Data Science
- Visualisation of both spatial and tabular data within Python
- Learn how to perform key GIS operations within the Python (geo-)data eco-system
- Exposure to state-of-the-art work on Integrating modern data science tools and techniques, and more standard GIS functionality
Dr Dani Arribas-Bel is a lecturer in Geographic Data Science at the Department of Geography and Planning and member of the Geographic Data Science Lab at the University of Liverpool (UK). Dani is interested in computers, cities and data. In particular, his work focuses on the spatial dimension of cities, from their physical structure to how socio-economic phenomena are spatially distributed. From a methodological standpoint, Dani is interested in incorporating new forms of data becoming available into the study of cities, as well as in computational methods such as spatial statistics and machine learning. Dani regularly teaches Geographic Data Science and Python courses at the University of Liverpool, and is a member of the development team of PySAL, the Python library for spatial analysis.
Who is this course suitable for?
The course will provide practical support to researchers and practitioners by introducing them to useful strategies to learn more from their spatial data. There is no expectation that delegates will have any previous knowledge of Python.