Unit information: Geographic Data Science in 2019/20

Unit name	Geographic Data Science
Unit code	GEOG30021
Credit points	20
Level of study	H/6
Teaching block(s)	Teaching Block 1 (weeks 1 - 12)
Unit director	Dr. Wolf
Open unit status	Not open
Pre-requisites	GEOG25010 Spatial Modelling 2
Co-requisites	None
School/department	School of Geographical Sciences
Faculty	Faculty of Science

Description including Unit Aims

This unit aims to provide students with an understanding of the methods, techniques, concepts, and theoretical-conceptual grounding for modern data science topics. The unit is structured as a methods survey unit, involving instruction in the concepts & theories behind geographical research in data science, as well as its applications in spatial supervised & unsupervised learning methods. The unit will cover three topics selected based on student interest including, but not limited to: (1) multilevel regression models; (2) spatial regression models; (3) local regression models; (4) generalised linear models; (5) spatial anomaly detection; (6) spatial clustering and regionalisation.

Intended Learning Outcomes

On completion of this Unit students will be able to:

1. Explain how spatial thinking is incorporated or embedded in a given spatial data scientific method or approach;
2. Estimate or run the methods covered in the course using R;
3. Understand and explain the substantive results of the methods in both technical and non-technical terms, either in writing and in presentation.

The following transferable skills are developed in this Unit:

• Numeracy, computer and problem solving;
• Analytical and quantitative skills and project management;
• Written and verbal communication

Teaching Information

Two 2-hour sessions per week, involving a combination of lecture and lab/practical work.

A common pattern of lecture/practical is generally adopted throughout the course:

• Introduction of concept, usually through graphs with a specific example
• Discussion of the formalisation of the concept in mathematical expressions
• Computation or estimation in R
• Interpreting & displaying the results, including visualisation & statistical display
• Directed reading of research papers applying the methods covered

Assessment Information

Coursework (60%) - ongoing practical assignments marked on best 3 out of 4. Practicals will be held weekly, and accumulated material will be submitted for assessment every 2 weeks. [ILOs 1-3] (Lab assignment 5 cancelled)

Take-home assessment (40%) - Over a 72-hour period, students will undertake a short analysis project using one method covered in the course. The example analysis will take the form of a journal article, report, or engineering blog post and will have a maximum length of 1200 words, not including bibliography or technical annexes. [ILOs 1-3]

Reading and References

The reading will focus on selections from books, two articles, as well as course notes written by the tutor. The following resources are an example of the material that will be covered in reading throughout the course:

• Baumer, B., D. Kaplan, & N. Horton.(2017) Modern Data Science with R, first edition. CRC Press, Boca Raton, FL.
• Hastie, T., R. Tibshirani, & J. Friedman. (2009) Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2^nd edition. Springer-Verlag, New York, NY.
• Gelman, A., & Hill, J. (2009) Data Analysis using Multilevel & Hierarchical Models. Cambridge University Press.
• Jones, K & Duncan, C. (2001) 'Using multilevel models to model heterogeneity: potential and pitfalls', Geographical Analysis, 32, (pp. 279-305) ISSN: 0016-736
• Jones, K. (2011) 'An introduction to statistical modelling', in Somekh,B Lewin,C (Eds.), Research methods in the social sciences, (pp. 236-250), Sage, 2011. ISBN: 0761944028
• LeSage, J. & R. K. Pace. (2009) Introduction to Spatial Econometrics. CRC Press, Boca Raton, FL.