Informatics 2 - Foundations of Data Science

INF2-FDS: Course team

mcorey — Tue, 08 Aug 2023 09:02:25 +0000

INF2-FDS: Course team mcorey Tue, 08/08/2023 - 10:02

The full list of people teaching and supporting FDS are on the Course Contacts page in Learn. Here we introduce the course lecturers and teaching assistants.

David Sterratt is a Lecturer in the School of Informatics, with research interests in computational neuroscience - understanding how parts of the brain and the central nervous system work by modelling the physics and biochemistry of cells and tissue. He is co-author of a textbook on computational neuroscience and maintains a number of R packages. In his spare time he helps to run a community news website, and listens to the excellent BBC More or Less programme to keep up to date with data science in the news.

Kobi Gal is a Reader in the School of informatics. He is interested in all that is Artificial Intelligence for the social good, especially in the design of intelligent tools to help students learn and teachers to understand and support their learning. He loves hill-running and running in Scotland.

Michael Gutmann is faculty in machine learning at the School of Informatics. He develops methods for inference and experimental design, and applies them in interdisciplinary work in the natural sciences.

Anna Hadjitofi is a Teaching Assistant on the course. She is a PhD student in the School of Informatics, interested in the neural circuits underlying honeybee communication. She has worked in several data scientist positions prior to her PhD and is looking forward to contributing to the course, especially during the final projects in the second term. In her spare time, she loves cycling around Edinburgh and East Lothian.

Narjes Rohani is a Teaching Assistant on the course. She is also a PhD candidate in Precision Medicine, where she applies data science algorithms to analyse the behaviour of Health Data Science students to facilitate HDS education. Prior to her PhD, she worked on several projects involving the use of Data Science to analyse biological data. In her spare time, she takes photographs, watches movies and writes poetry.

License

INF2-FDS: Resource List

flittlet — Tue, 18 Jul 2023 14:48:03 +0000

INF2-FDS: Resource List flittlet Tue, 18/07/2023 - 15:48

Lecture notes

The main reading for the course is the FDS lecture notes.

FDS-lecture-notes-2024-04-18.pdf (6.46 MB)

Please email david.c.sterratt@ed.ac.uk if you would like the lecture notes in a different format.

There is also other essential and recommended reading on the Resource List below.

Informatics 2 - Foundations of Data Science Resource List

You can access the reading list for this course by selecting the link above. In order to view some resources on the list, you must be logged in with your EASE account.

For more information on getting the most out of your courses Resource List, have a look at this video.

License

INF2-FDS: Course Materials

flittlet — Tue, 18 Jul 2023 14:47:46 +0000

INF2-FDS: Course Materials flittlet Tue, 18/07/2023 - 15:47

Overview of course content

The Schedule gives an overview of the course.

The lecture notes cover much of the course content. Each chapter corresponds closely to one or more lectures, as indicated in the Schedule, and we recommend that you read the relevant chapter before each lecture.

FDS-lecture-notes-2024-04-18.pdf (6.46 MB)

As the semester goes on, we'll place course materials in the subpages linked from this one. Materials include:

Lecture slides
In some weeks, Jupyter notebooks with examples shown in the lecture
Lab notebooks for that week's Lab - including exercises, hints you can reveal if you get stuck and solutions you can reveal to check your answers.
Reminders about Comprehension Questions in Learn Ultra/Gradescope which are designed to check and develop your understanding of the lecture material and give you feedback on how well you have understood the concepts. They do not form part of the course mark, and you can repeat them as often as you like.
Tasks for you to do in preparation for the next week's workshop

Lecture Recordings

All lecture recordings should be accessed via Learn Ultra; you will need to log in using your EASE account. (Learn Ultra provides you with access to any lecture recordings available for this course. You will need to select the "lecture recording" link once, before you can access any direct links to a lecture recording.)

License

INF2-FDS: Informatics 2 - Foundations of Data Science

flittlet — Tue, 18 Jul 2023 14:47:45 +0000

INF2-FDS: Informatics 2 - Foundations of Data Science flittlet Tue, 18/07/2023 - 15:47

Welcome to Informatics 2 - Foundations of Data Science

We, the FDS course team are looking forward to helping you learn about the Foundations of Data Science. This OpenCourse site is the main source for most course information, including:

an overview of what expect you to learn and how you'll be assessed
the schedule
lab and workshop logistics
course materials, including the lecture notes and slides.

Our course Learn page will still be used for communicating important messages (via the Announcements feature), for coursework information and submitting coursework.

About the course

The amount of data is growing at an exponential rate, and it has the power to transform our world for better or worse. It's crucial that we learn how to collect, process and present data so that the conclusions we draw from the data are well-founded, and that we can communicate the results clearly both graphically and in writing. It's vital that we understand the legal frameworks that govern issues around data use, in particular privacy and licensing. We also need to go beyond the law, and consider ethical issues: for example, do we have the consent of the people providing the data in a survey and are the conclusions we draw valid? It's also fundamental that we understand how algorithms we use work, and how much confidence we should have in our results - which takes us into the realm of statistics.

In this course, we'll introduce these foundations of data science, using real world examples of data sets and ethical scenarios. You'll lean how to use Python libraries such as Pandas, Matplotlib and Seaborn to undertake data wrangling and visualisation. We'll introduce linear models both for explanation and prediction, and we'll give a taster of some machine learning methods, which are covered in more depth in later courses in Informatics degree programmes.

License

INF2-FDS: Schedule

flittlet — Tue, 18 Jul 2023 14:47:45 +0000

INF2-FDS: Schedule flittlet Tue, 18/07/2023 - 15:47

Timetable

The course has various learning activities, which are coordinated with each other and the assessment. The timetables are a bit different for Semester 1 and Semester 2:

Weekly activities

We recommend that each week you have a pattern of:

Doing the reading listed before the lectures. Most of the reading is from the FDS lecture notes:

FDS-lecture-notes-2024-04-18.pdf (6.46 MB)

Attend the lectures, which include exercises, discussion, Q&A, demos or feedback on exercises. The lectures are delivered by Kobi Gal (KG), David Sterratt (DS) and Michael Gutmann (MG).
Test yourself with the comprehension questions in Learn Ultra afterwards.
Do the lab notebook - in Weeks 1 to 7 there will be lab sessions in Appleton Tower with demonstrators; after then the labs are self-study
Attend the workshops - preparation the week before is ideal, but if you've not managed to prepare, you should get something from the workshops. Two of the workshops are designed to familiarise you with the coursework released shortly after, including how we mark it.

Coursework

There are 3 courseworks and one class test; you can see the schedule in Learn Ultra or import the data into your calendar and use the coursework calendar file https://course.inf.ed.ac.uk/calendar/inf2-fds.ics to get all the coursework events into your Outlook or Google calendar.

Schedule

Week	Lecture 1	Lecture 2	Lab	Task/workshop	Reading
Data: ethics, collection, representation, wrangling, exploration, visualisation and descriptive statistics
S1 W1 18-22 Sep	Introduction and Logistics (KG)	Data (KG)	Introduction to Jupyter notebooks and Pandas		Lecture Notes (LN) 1 and 2
S1 W2 25-29 Sep	No lecture	Descriptive statistics (KG)	Pandas - Data wrangling	Task: Preparation for Week 3 Workshop on Ethics.	LN 3, An Introduction to Data Ethics, Parts 1 and 2
S1 W3 2-6 Oct	Exploratory data analysis, data communication visualisation (DS)	Visualisation (NR)	Data representation I - Matplotlib	Workshop: Data ethics discussion	LN 5, The Big Book of Dashboards, Chapter 1
S1 W4 9-13 Oct	Intro to data ethics (KG)	Data collection and statistical relationships (KG)	Data representation II: Distributions	Task: Marking visualisations and interpretation in previous CW1	LN 4
Introduction to Machine Learning
S1 W5 16-20 Oct	Intro to supervised learning: Nearest neighbours (KG)	k-Nearest Neighbours and Evaluation (KG)	Data wrangling II	Workshop: Discussing visualisations and interpretation in previous CW1	LN 7
Linear Models
S1 W6 23-27 Oct	Linear regression I (MG)	Linear regression II (MG)	k-Nearest Neighbours		LN 10
S1 W7 30 Oct-3 Nov	No lecture	No lecture	No lab
S1 W8 6-10 Nov	Multiple regression I (MG)	Multiple regression II (MG)	Linear models	Task: Preparation for Week 9 workshop	LN 11 and 12
S1 W9 13-17 Nov	Principal Components Analysis I (MG)	Principal Components Analysis II (MG)	PCA	Workshop: Interpretation of data science study using multiple regression	LN 13
Statistical inference
S1 W10 20-24 Nov	CW1 feedback and intro to inferential statistics (DS)	Randomness, sampling and simulation (DS)	Randomness, sampling and simulations		LN 14
S1 W11 27 Nov-1 Dec	Estimation (DS)	Confidence intervals (DS)	Estimation of confidence intervals with the bootstrap		LN 15 and 16
S2 W1 15-19 Jan	Hypothesis testing and p-values (DS)	A/B testing (DS)	No lab	Task: Preparing for CW2	LN 17 and 18, XKCD comic strip on multiple testing, A hypothesis is a liability
S2 W2 22-26 Jan	Logistic regression (DS)	Logistic regression and CW2 Q&A	Logistic regression	Workshop: Preparing for CW2	LN 19
Introduction to Machine Learning
S2 W3 29 Jan-2 Feb	Intro to unsupervised learning: K-means (DS)	No lecture		Task: Problem sheet for S2 Week 4 Workshop	LN 9
S2 W4 5-9 Feb	Ethical issues with supervised learning (DS)	No lecture	K-means	Workshop: Statistical problems 1	Equality law can disadvantage women in algorithmic credit decisions
Regression and Inference
S2 W5 12-16 Feb	Linear regression and inference (DS)	Generalised linear models (DS)	Web-scraping	Task: Problem sheet for S2 Week 6 Workshop	LN 20
S2, FLW 19-23 Feb
Project and project skills
S2 W6 26 Feb-2 Mar	Software engineering for data science (AH+DS)	No lecture		Workshop: Statistical problems 2	Good enough practices in scientific computing , Reproducible Analysis Through Automated Jupyter Notebook Pipelines , The scientific paper is obsolete
S2 W7 5-9 Mar	Project Q&A (AH+DS)	No lecture
S2 W8 12-16 Mar	Project writing workshop (DS)	No lecture
S2 W9 19-23 Mar		No lecture		Workshop: Mid-project presentation (TA+DS)
S2 W10 26-30 Mar	No lecture	No lecture		Workshop: Mid-project presentation (TA+DS)
S2 W11 2-6 Apr

License