Geographic Data Science with PySAL and the PyData Stack¶
Sergio J. Rey
Levi J. Wolf
This book provides the tools, the methods, and the theory to meet the challenges of contemporary data science applied to geographic problems and data. Social media, new forms of data, and new computational techniques are revolutionizing social science. In the new world of pervasive, large, frequent, and rapid data, we have new opportunities to understand and analyze the role of geography in everyday life. The book provides the first comprehensive curriculum in geographic data science.
Geographic data is ubiquitous. On the whole, social processes, physical contexts, and individual behaviors show striking regularity in their geographic patterns, structures, and spacing. As data relating to these systems grows in scope, intensity, and depth, it becomes more important to extract meainingful insights from common geographical properties like location, but also how to leverage topological properties like relation that are less commonly-seen in standard data science.
This book introduces a new way of thinking about geographic challenges. Using geographical analysis and computational reasoning, it shows the reader how to unlock new insights hidden within data. The book is structured around the excellent data science environment available in Python, providing examples and worked analyses for the reader to replicate, adapt, extend, and improve.
Geographic Data Science is a new field, but has many academic influences and precursors. Currently, curriculum development is forced to combine snippets of unpublished research code and vignettes together with various packages in the Python data science ecosystem. This book considers students as learners of both analytical and computational methods. We provide accessible, open computational examples with high-quality narrative exposition. This combination is design to satisfy the two main requirements from students:
As learners of analytical methods, students want the narrative scaffolding, careful explanation, and intelligible writing provided by typical introductory textbooks.
As learners of computational approaches, students need worked examples of code alongside an explanation on how to get started even doing analysis.
There are many hurdles over which students jump just to start the first example of many introductory textbooks. Our book provides a full view of Geographic Data Science, from setting up and organizing computational environments to preparing data, through to developing novel spatial insight and presenting it cogently to others. This kind of integrated approach is necessary for data science, and is particularly relevant in geographical settings where mapping is of central importance.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.