Data Science MSc

You develop your ability to design and implement database, big data and analytics applications to meet business needs. A case study is used to follow the system development lifecycle. You develop a plausible application from inception to implementation for a real-world scenario.

You investigate the issues and technologies associated with implementing and supporting large scale databases and the services that are needed to maintain and access a repository of data. Investigations are undertaken in a number of areas including big data, data warehouses, integrating legacy data, data management and approaches that support the modelling and visualisation of data for a range of use views.

You undertake a major, in-depth, individual study in an aspect of your course. Normally computing master’s projects are drawn from commercial, industrial or research-based problem areas. The project involves you in researching and investigating aspects of your area of study and then producing a major deliverable, for example software package or tool, design, web-site and research findings. You also critically evaluate your major deliverable, including obtaining third party evaluation where appropriate.

The major deliverable(s) are presented via a poster display, and also via a product demonstration or a conference-type presentation of the research and findings. The research, project process and evaluation is reported via a paper in the style of a specified academic conference or journal paper. The written report, the major deliverable and your presentation of the product are assessed.

The project management process affords supported opportunities for goal setting, reflection and critical evaluation of achievement.

The field of information visualisation has expanded rapidly with many designers generating new forms of charts through which to view quantitative data. This module explores the range of charts available from the traditional such as bar charts and pie charts, to the more novel such as stream graphs, tree maps, sunbursts, and force diagrams, and examines their mathematical properties.

By accurately representing quantitative data using appropriate charts, the intended audience can make their own interpretations of the data and identify emerging patterns and themes that are more readily recognisable in chart form than in the form of raw data.

Dynamic, interactive visualisations enable the reader to explore the data for themselves through a variety of perspectives. Static visualisations are excellent for print medium but are restricted to showing a single perspective and do not handle multidimensional datasets well. Using an interactive graphic the reader can zoom in on sections of the data which are of interest, explore more than one dimension at a time, and sort and filter to discover new patterns and themes within the data. Particularly useful is the ability to provide a macro/micro view of the same data, ie a big picture view of the full dataset from which the reader can then ‘drill down’ into the lower level detail.

This module uses the javascript library for Data-Driven Documents (D3js) for creating animated, dynamic graphics for the web, and looks at other alternatives available.

Machine learning is a subfield of computer science concerned with computational techniques rather than performing explicit programmed instructions. You build a model from a task based on observations in order to make predictions about unseen data. Such techniques are useful when the desired output is known but an algorithm is unknown, or when a system needs to adapt to unforeseen circumstances.

You explore statistics and probability theory as the fundamental task is to make inferences from data samples. The contribution from other areas of computer science is also essential for efficient task representation, learning algorithms, and inferences procedures. You gain exposure to a breadth of tasks and techniques in machine learning.

Assessment is an in course assessment (100%).

You develop the knowledge and skills to understand the research process in the field of computing and gain the necessary skills to undertake your masters project. You learn how to evaluate previous academic research and generate evidence material to justify your research. You learn different methods of data generation and develop an understanding of how these methods fit into your primary research, development lifecycle, evaluation of the end user experience, use of academic research literature and research ethics.

You develop necessary knowledge and practical understanding of the main statistical techniques. You explore quantitative and qualitative data analysis techniques, reflecting scientific and social science methods. You focus on correlation testing, regression, data categories, normalization - the tools needed, rather than the philosophical approaches. You understand how to apply valid techniques and interpret the results in preparation for experimental work.

Your assessment is a single ICA based around a number of case studies that require you to identify the correct data analysis and modelling processes.

The internship options are:

Vocational: spend one semester working full-time in industry or on placement in the University. We have close links with a range of national and international companies who could offer you the chance to develop your knowledge and professional skills in the workplace through an internship. Although we cannot guarantee internships, we will provide you with practical support and advice on how to find and secure your own internship position. A vocational internship is a great way to gain work experience and give your CV a competitive edge.

Research: develop your research and academic skills by undertaking a research internship within the University. Experience working as part of a research team in an academic setting. Ideal for those who are interested in a career in research or academia.