Postgraduate Physics 2019-2020

Data Science is an interdisciplinary field that seeks to identify and understand phenomena captured in structured or unstructured data, extract insights, and add value by generating predictions that aid optimization of processes and equipment. These techniques show considerable promise for bringing about a revolution, increasing the significance and value of owning and collecting data of all types. This course introduces the common techniques and considers the implications for data managers.

This course teaches programming in high level languages and in particular the Wolfram Language (Mathematica). It will introduce all areas of this powerful language, including symbolic and numerical calculations and simulations, links to other high level languages such as R and Python, links to database languages mySQL and Mongo.

We will show how Wolfram Language allows computation to be applied to many areas of data analysis, and modelling. This allows us to gain deep insight into systems.

In this course we study the typical workflow for a data analysis project. We will learn how to access and collect data, how then to clean the data, and organise it in databases to prepare it for later analysis.

We will then perform descriptive and exploratory data analysis and finally visualise the results and create a report.

In this course we will discuss modern methods of machine learning, such as decision trees, regression, Markov models, Bayesian approaches, Nearest Neighbours, random forests, support vector machines and neural networks.

Great emphasis will be given to the actual application of all these methods to small and large data sets.

This is an introductory course in statistics and statistical methods for data analysis.

We will introduce descriptive statistics, ANOVA, GLMs, correlations, spectra, wavelets, etc.

This will allow us to perform typical analysis that underlie most modern data science questions.

Visualising the outcome of a data analysis is critical to communicate the results. In this course we will study standard and cutting edge visualisation techniques to make sense of data, and present it in a compelling, narrative-focused story.

Presenting and visualising data and reporting on the result of an analysis are a crucial skill when making sense of data.

In this module we will discuss advanced and cutting-edge statistical tools and techniques.

Some of the topics covered are likelihood, advanced hypothesis testing, outlier detection, data imputation, bootstrap, nonparametric regression and mixed effect models.

This course introduces the tools needed to analyse audio recordings, images and videos.

It will contain aspects of image enhancements, content detection, segmentation analysis (e.g. detecting tumours in medical imaging data), handwritten character recognition, subtitle analysis, and many other techniques.

This course brings together all aspects of data science, from gathering data, to analysis and visualisation, by exploring real world applications of data science. There will be a discussion of some significant achievements of data science when applied to various areas from fundamental business practices to physics. Students will then apply these skills to a group project.

This is a project course for the MSc in data science. Students will be given a data science project, which will be supervised by two members of staff.

The project will involve a typical data science workflow, from data collection, cleaning, to analysing and visualising the results.

The students will have to deliver a presentation and hand in a report about the results.