Dr Neil Olver

This course is available on the MSc in Applicable Mathematics, MSc in Marketing and MSc in Operations Research & Analytics. This course is available as an outside option to students on other programmes where regulations permit.

This course has a limited number of places (it is controlled access). Priority is given to MSc students in the Department of Mathematics.

Students are not permitted to take this course alongside ST443, Machine Learning and Data Mining.

Students must have knowledge of Statistics and the programming language R to the level of ST447, Data Analysis and Statistical Methods, or alternatively, a comparable knowledge of Python.

The course introduces fundamental machine learning methods for data analytics problems. Vast quantities of data are available today in all areas of business, science, and technology as well as social networks. The goal of data mining is to extract useful information from massive-scale data. The aim of this course is to equip students with theoretically grounded and practically applicable knowledge of the most important machine learning algorithms used for this task, as well as how they should be applied. Mathematics (e.g., optimisation, graph theory), computer science and statistics all play an important role.

For classification and regression problems, methods studied include naive Bayes, K-nearest neighbours, decision trees, support vector machines, and neural networks. The course will also cover unsupervised learning methods such as clustering. Ethical issues arising from machine learning are also discussed.

The methods are illustrated on practical problems arising from various fields. Students will make use of various machine learning and data mining packages in R and Python, as appropriate.

This course is delivered through a combination of seminars and lectures totalling a minimum of 30 hours across Winter Term.

There will be a formative group project, in preparation for a similar summative project.

• James, Witten, Hastie, Tibshirani, An Introduction to Statistical Learning: with Applications in R (2016)
• Hastie, Tibshirani, Friedman, The Elements of Statistical Learning: Data Mining, Inference and Prediction, 2nd ed. (2009)
• Witten, Frank, Hall, Data Mining: Practical Machine Learning Tools and Techniques, 3rd or 4th ed. (2016)

Exam (60%, duration: 2 hours) in the spring exam period.
Project (40%) in the ST.

The examination is critical to assessment. In order to pass this course, students need to achieve a mark of at least 50% in the examination. A fail mark in the exam will result in an overall fail mark for the course: it cannot be compensated by the marks in the other elements.