Date/Time
Date(s) - 25/03/2018 - 29/03/2018
All Day
Categories No Categories
The course will provide an introductory overview to several key concepts and tools behind the process of doing Data Science. We will cover topics from data manipulation and visualization, to exploratory data analysis, to learning from models. The course will be taught entirely in Python, the modern industry standard for data science, and will have a substantial hands-on component
Course Contents:
This course will introduce the participants to the process of doing Data Science. This covers all the steps involved in solving practical problems with data: design, manipulation, exploration, and modeling, as well as learning from models. These topics will be explored from a “hands-on” perspective using a modern Python stack (e.g. pandas, seaborn, scikit-learn, PySAL), the industry standard, and examples from real-world spatial and tabular data.
We will spend time reviewing recent workflows suggested to obtain (e.g. APIs) and reshape (e.g. the “tidy data” paradigm; Wickham, 2014) data from disparate sources. Then we will move on to techniques to visualise and summarise your data, including unsupervised learning algorithms for clustering. From there we will cover modeling data and discuss the different perspectives that the statistics and machine learning communities provide. We will end by discussing ways to evaluate and learn from predictive models you have built. The course is intended to provide practical support to researchers and practitioners by introducing them to useful strategies to learn more from their data. For this reason, participants are encouraged to bring their own datasets and problems as there will be time and space to discuss them.
Day 1 – Introduction
- Data, data, data: the rise of new forms of data
- What is Data Science?
- The modern Python stack for Data Science
Day 2 – Exploring data
- Data plumbing: ingestion, manipulation, presentation
- Visualisation: tabular and spatial data
- Unsupervised machine learning – Example: clustering with K-means
Day 3 – Machine learning
- Supervised learning – Example: regression
- Learning from models
Day 4 – Data science studio
- Hands-on data dive
- One-on-one tutorials
The course is introductory and, as such, it will provide a panoramic overview of several concepts and techniques. Basic statistical notions as well as some experience with programming are not strictly required but will be helpful.
Contacts:
Please contact the following people if you require additional information:
Becca Prescott – soesms@liverpool.ac.uk
Dr Dani Arribas-Bel – D.Arribas-Bel@liverpool.ac.uk
Course Fees:
- £400 students and academics (w/ University affiliation)
- £700 other
One day only:
- £140 students and academics (w/ University affiliation)
- £200 other
Cancellation:
A full refund can be given for any cancellations received more than 7 days before the start of the course. Any cancellations received 7 or fewer days before the course will be charged at the full rate.
Catering and Learning Preferences:
Please make any catering or learning preferences known to Becca Prescott when booking the course by emailing soesms@liverpool.ac.uk or by telephone 0151 794 3085.