Spatial Data Science

This class will train students to gather, merge and clean data from multiple sources. We will focus on urban areas with a spatial lens to gain valuable insights into the reality of multiple societal problems like climate change, energy transition, and inequalities. With unprecedented growth in cities, policymakers strive to balance providing equal opportunity and benefit to its citizens and sustainable development. Data science will help us understand and estimate multiple implications of solutions and communicate results to a broad audience effectively.

The course is divided into five major modules, each focusing on crucial steps in the lifecycle of a data science project.

• Obtain : Obtaining data from multiple [open] data sources.

• Scrub : Data cleaning, munging, and sampling to consolidate all information into a dataset that is manageable, informative and relates to your problem.

• Explore : Exploratory data analysis to make sense of what your data is trying to say.

• Model : Estimation and modelling based on statistical tools such as regression, classification and clustering.

• Interpret : Communicating results and reflections through visualisation, storytelling and interpretable summaries.

Students will need their laptops only during labs. All classroom sessions will be a mixture of interactive lectures, discussions and labs. Labs will cater to a series of exercises to reinforce the skills and topics presented in class. A final graded group project (group size 4 students) will allow students to run through the entire cycle of the five modules: obtaining, cleaning, analyzing, interpreting, and communicating data. As the final deliverable, each group will produce an article that documents their work across all five modules, presenting their chosen research problem, methods, analyses, results, and conclusions. The article will be graded based on criteria including clarity of research question, appropriate data handling and analysis, quality of interpretation, and effectiveness of communication.

For students who have had statistical, math or computer programming courses in their bachelor's or elsewhere, this course will add to your skills by providing you with tools to become future policy-makers, academics, responsible data scientists, and supporters of open science.

The course has 8 contact hours a week, divided into two interactive and blended lecture hours, 4 lab hours and two open hours to ask questions weekly. Although attendance for the contact moments is not mandatory, substantial self-study will be required to keep up with the course if you do not attend at least 4 hours a week.