Master of Data Science (Earth and Environment)

As part of a suite of Data Science Masters this course is targeted at Geographers, Earth and Environmental Scientists who want to learn how to work with managing natural resources and spatio-temporal information flows. These courses are designed to equip students with the key skills and interdisciplinary knowledge to address the challenges of utilising the streams of data produced in modern industry, science and government.

The advantage of taking one of these Masters at Durham is that you will be part of a cohort of Masters students who will work together throughout the 12 month course. Included with the Masters course your research project could be carried out in conjunction with a range of partners across industry, regulators and policy-makers building on Durham’s significant strengths in delivering world-leading and world changing research.

To equip you for your career all of the students on the suite of Data Science Masters courses share core modules. These shared modules mean you will be carrying out team building activities, present case studies and carry out both formative and summative assessments with students from all four faculties of Durham University ensuring that you learn how to represent not just your own discipline but listen and the integrate views and skills from other disciplines.

All around us, massive amounts of increasingly complex data are being generated and collected, for instance, from mobile devices, cameras, cars, houses, offices, cities, and satellites. Business, research, government, communities, and families can use that data to make informed and rational decisions that lead to better outcomes. It is impossible for any one individual or group of individuals to keep on top of all the relevant data: there is simply far too much. Data science enables us to analyse large amounts of data effectively and efficiently and as a result has become one of the fastest growing career areas.

Previously, data science was the province of experts in maths and computer science, but the advent of new techniques and increases in computing power mean that it is now viable for non-experts to learn how to access, clean, analyse, and visualize complex data. There is thus a growing opportunity for those already in possession of knowledge about a particular subject or discipline, and who are therefore able to grasp the full meaning and significance of data in their area, to be able to undertake data analysis intelligently themselves. The combination of primary domain knowledge with an expertise in extracting relevant information from data will give those with this ‘double-threat’ a significant employment advantage.

The Master of Data Science programme is a course with a hard-core of data science, intended to provide Masters-level education rich in the substance of data science for students who hold a first degree that is not highly quantitative, including those in social sciences, the arts and humanities. Introductory modules are designed to bring students with non-technical degrees up to speed with the background necessary for data science. This is done on a need-to-know basis, focusing on understanding in practice rather than abstract theory. Core modules then introduce students to the full range of data science methods, building from elementary techniques to advanced modern methods such as neural networks and deep learning. Optional modules allow students to focus on an area of interest.

The programme provides training in relevant areas of contemporary data science in a supportive research-led interdisciplinary learning environment. The broad aims are:

• To develop advanced and systematic understanding of the complexity of data, including the sources of data relevant to science, alongside appropriate analysis techniques
• To enable students to critically review and apply relevant data science knowledge to practical situations
• To develop a critical awareness of current issues in data science which is informed by leading edge research and practice in the field
• To develop a conceptual understanding of existing research and scholarship to enable the identification of new or revised approaches to data science practice
• To develop creativity in the application of knowledge, together with a practical understanding of how established, advanced techniques of research and enquiry are used to develop and interpret knowledge in data science.
• To develop the ability to conduct research into data science issues that requires familiarity with a range of data, research sources and appropriate methodologies and ethical issues
• To develop advanced conceptual abilities and analytical skills in order to evaluate the rigour and validity of published research and assess its relevance to new situations
• To extend the ability to communicate effectively both orally and in writing, using a range of media.

The programme is designed around a pedagogical framework which reflects the core categories of the data science discipline.

A number of subjects can be identified and defined within each application domain. Whilst a Masters programme cannot incorporate all subjects, a selection of subjects representative of each domain ensures that the programme incorporates the necessary breadth and depth of material to ensure a skilled graduate.

The programme allows for progressive deepening in the students’ knowledge and understanding, culminating in the research project which is an in-depth investigation of a specific topic or issue.

The global dimension is reinforced through the use of international examples and case studies where appropriate.

Course Structure

Core modules:

• Introduction to Computer Science (optional under certain conditions)
• Introduction to Statistics for Data Science
• Machine Learning
• Programming for Data Science
• Data Science Tools in Earth Sciences
• Data Science Applications in Earth Sciences
• Introduction to Maths for Data Science (optional under certain conditions)
• Research Project

Examples of optional modules:

• Text Mining and Language Analytics

• Data Exploration, Visualization, and Unsupervised Learning