Teaching

The taught element comprises two module types; major modules which cover all essential aspects required for scientific data analysis, and minor modules which demonstrate the application of these data science skills to real-world scientific research areas. Candidates will be required to take seven modules in total for assessment comprising 5 major and 2 minor modules. These will be chosen from:
•    a selection of (minimum 6) Major modules
•    a selection of (minimum 4) Minor modules

The major modules will be taught in Michaelmas and Lent terms. They will cover the main material that all students are expected to master covering three topic areas: statistical data analysis, machine learning and AI , and research computing, high performance computing, and software development. These major modules will equip students with the essential skills for top-level scientific data analysis.

The minor modules will be taught in Lent and Easter terms. The selection of minor modules offered will be updated annually to respond to changes in the research landscape. These will cover the application of the core techniques taught in the major modules to specific scientific research problems. Beyond illustrating useful data science methodology that may have broader applications, the minor modules are also intended to give insight into the goals and status of the relevant field and hence can help in preparing interested students for research in one of these areas.

The data analysis project will focus on investigating the reproducibility of a key scientific data analysis in the literature. Students will select their project from a pre-approved list in Michaelmas term and will work on it over Lent and Easter terms with submission at the end of the course. The projects will be designed to be open-ended so students can improve or extend what is already published in the literature.  Several students will be able to select each project topic, but they will work independently and will be offered individual supervision with the responsible academic advisor, as well as further advice and support from the teaching team.  

The programme will also feature non-assessed content designed to broaden students' knowledge.  This may take the form of short modules on specialist topics, communication workshops, a weekly seminar series and modules on industrial applications led by our industry partners.

Feedback

Feedback on student performance on their written examination, oral presentation and coursework results is provided by the module leader; feedback on the data analysis project progress is provided by their data analysis project supervisor.

Postgraduate students are represented on the Department's Postgraduate Student Consultative Committee, which normally meets five times a year, and consists of one or more student representatives from each of the research groups or master’s degree programmes. The Committee exists to enable discussion of any issue affecting postgraduate studies and students may approach any member of the Committee to suggest items for discussion.

Assessment

Thesis / Dissertation

Data Analysis Projects will primarily be concerned with the reproducibility of key scientific analysis. Projects will be marked on three aspects: the project reports, the accompanying data analysis pipeline developed for the analysis and the oral presentation of the project.  

The report must not exceed 7,000 words in length and describe the analysis pipeline and its development, and the project goals and results obtained.  The report must be accompanied by an executive summary of the work of not more than 1,000 words in length.

The data analysis pipeline used for the analysis must also be provided to the assessors in a form which is accessible and reproducible.

The oral presentation will be used to confirm the candidates understanding of the project and to clarify any points which were unclear in the report or analysis pipeline.  Assessors may ask questions of the candidate during the presentation to further explore any aspect of the project, the submitted materials, the presentation, or other background knowledge relevant to the project.

Other

Each Major and Minor Module will be assessed via a mix of:

Coursework will typically be in the form of a report describing the development and implementation of specific data analytic methods, typically of not more than 3,000 words in length, in conjunction with the data analytic pipeline itself, however the exact form will be module dependent.  The reports will be expected to be concise and will be judged on the quality of argument, the clarity of presentation, and the insightfulness of interpretation.  The pipeline itself will be judged on conformity to software development best practice as taught in the 'Research Computing' major module, and the quality of the pipeline in terms of its accuracy, range of application, ease of use, and robustness and stability.

Written exams will be closed books and will primarily test candidates' theoretical knowledge via calculations, short answer questions and essays.

Oral presentations on candidates’ work may be required as part of the assessment of submitted coursework.

In the MPhil in Data Intensive Science, the weighting of the assessed course components is as follows:

1. the Project report (Data Analysis project) will represent 25 per cent (25%) of the final grade;
2. the taught modules examination (mix of written assignment, written examination, and oral presentation) will represent 75 per cent (75%) of the final grade where :
   • Each major module will count for 12% of the final grade.
   • Each minor module will count for 7.5% of the final grade.