MSc Data Science

Programme Specification
Awarding Institution:
University of London (Interim Exit Awards made by Goldsmiths’ College)
Teaching Institution: Goldsmiths, University of London
Name of Final Award and Programme Title: MSc Data Science
Name of Interim Exit Award(s): Postgraduate Diploma in Data Science
Duration of Programme: 1 year full-time or 2 years part-time
UCAS Code(s): Not applicable
HECoS Code(s): (100366) Computer Science
QAA Benchmark Group: Computing
FHEQ Level of Award: Level 7
Programme accredited by: Not applicable
Date Programme Specification last updated/approved: August 2017
Home Department: Computing
Department(s) which will also be involved in teaching part of the programme:
Not applicable

Programme overview
Computer Science is one of the main research areas in the Department of Computing at
Goldsmiths, and is a key component of the multiple interdisciplinary research directions, in
particular of Data Science, which are developed within the Department. Both the science of
efficient computing and the science of extracting information from data continue to increase
in importance in various disciplines in which the large volume and the complexity of data
used in research impose unprecedented challenges to the data analysis approaches
traditionally employed in these disciplines. Tractable and efficient solutions are needed for
big data management and analysis nowadays, and Data Science is the answer.

The Data Science MSc programme at Goldsmiths, which is built upon the existing expertise
in Statistical Computing, Data Mining, Machine Learning, Time Series Forecasting, Soft
Computing, Big Data, Algorithmics and Databases in the Computing Department, responds
to the increasing need for applying modern computing approaches to data processing and
analysis in new monodiciplinary and interdisciplinary research in the Department and the
College, and for providing students with postgraduate training in this area.

MSc Data Science - Programme Specification 1

Goldsmiths, University of London
The programme is well integrated in the multidisciplinary teaching and research activity in
the College. In research terms, the Data Science MSc programme will complement the
research group in Data Science and Soft Computing that students can join for their final
project or for PhD studies subsequently. Students will equally have the opportunity to join
other research teams in the College for their final project work, in which they can contribute
with their gained expertise in Data Science to the interdisciplinary work developed by those
teams.

In particular students will be exposed to such research work in Data Science or to

multidisciplinary research work employing Data Science via the Data Science Research
Topics 15-credit module in which researchers from across the College will present their
work and attract students' interest to possibly contribute to that work during a final project. In
the same module students will have the opportunity to attend presentations by professionals
specialised in Data Science, based on current common concerns and cutting-edge specific
technologies and applications used in industry.

Students will receive highly specialised training in various conventional and modern
statistical, machine learning and data mining algorithms for data analytics, and in highly
scalable methods and technologies used in the management and analysis of big data.
These topics form the basis of the two compulsory e 30-credit modules, namely Machine
Learning and Statistical Data Mining on one hand, and Big Data Applications on the other.
In particular students will learn these concepts and techniques through the use of major
software technologies for data analysis and mining, including text and web mining, such as
R, IBM SPSS Statistics and RapidMiner, and on big data software technologies as Hadoop
based on HDFS distributed file system and the MapReduce scalable computing approach,
NoSQL databases, Hive data warehousing, Pig Latin, as well as general-purpose
programming languages as Python. These software technologies will be running on a
dedicated computer cluster sustaining large applications.

Students will be exposed to real world applications by modelling and implementing

applications encountered in business (including customer analytics, credit scoring, financial
forecasting, reality mining, mobile telephone data analysis), in health and medical research
(including automatic diagnosing, genetic data mining and bioinformatics, mining online
medical publication libraries), in unstructured data analysis (including text and web mining
applied in sentiment analysis and the intelligent web). Further specialised training is
provided in elective modules which are shared with other programmes. These electives
include Neural Networks, Introduction to Natural Language Processing, Linked Data and the
Semantic Web, Interaction Design, Artificial Intelligence, Open Data, and Music Information
Retrieval (subject to timetabling and staff availability). These modules provide further
specialisation in various topics related to forecasting and soft computing, web mining,
information retrieval, intelligent systems, and open data, which are directly linked or
complement the Data Science topics studied in the compulsory modules of the programme.

MSc Data Science - Programme Specification 2

Goldsmiths, University of London
The Final Project module, worth 60 credits, will allow students to undertake substantial
independent projects that will allow them to demonstrate project planning and management
skills, research skills, and written and oral presentation skills. Students will integrate their
knowledge and will use skills acquired in the other programme's modules in the
implementation of their final project in Data Science or related interdisciplinary topics.

Programme entry requirements

Upper second class undergraduate degree or above in computing, mathematical sciences,
physics, engineering, finance, bioinformatics and computational biology, social sciences
with a good basis of statistical knowledge and some programming experience.

Prospective students should have an interest in one or more Data Science related topics
including statistics and data analysis, machine learning, data mining, big data,
bioinformatics, intelligent web, financial forecasting, and computational social sciences.

Outstanding practitioners or individuals with strong commercial experience may be

considered. Nonnative English students should normally have a minimum IELTS score of
6.5 or equivalent.

Aims of the programme

The aim of this Programme is to produce graduates who are autonomous, creative and
reflective computing practitioners, and have in particular practical skills and research
abilities in the Data Science field. Our graduates should have:

• Knowledge of computing technologies across a range of specialist topics in Data

Science, both in terms of the latest research advances and industry standards.
• The ability to design and implement from small to large scale data analysis
processes, by using existing specialised software solutions or by implementing their
own software solutions, on usual or specialised scalable computing hardware,
including computing clusters.
• Strong transferable skills, in particular the ability to work independently and in groups
and reflectively evaluate their own work.
• The ability to perform monodisciplinary research in Data Science and interdisciplinary
research in work involving also data analysis or other topics related to Data Science,
or the ability to work with industry. In particular our students are encouraged to join a
research team in the College, or undertake industry internships as part of their final
project.

MSc Data Science - Programme Specification 3

Goldsmiths, University of London
What you will be expected to achieve
Learning outcomes for the Postgraduate Diploma:

Knowledge and understanding

Code Learning outcome Taught by the following module(s)

A1 A deep and practical understanding of Data Programming, Big Data
state of the art of scalable computing Applications, Machine Learning and
technology particularly applied on Statistical Data Mining, Data Science
variable volumes of data Research Topics, Linked Data and
the Semantic Web, Open Data
A2 A systematic understanding of machine Machine Learning and Statistical
learning and statistical data mining Data Mining, Data Science Research
techniques used in data analytics and in Topics, Neural Networks, Introduction
other related areas to Natural Language Processing,
Music Information Retrieval
A3 A good understanding of various topics Data Science Research Topics,
involved in interdisciplinary research in Machine Learning and Statistical
which Data Science is applied Data Mining, Final Project, Interaction
Design, Music Information Retrieval
A4 A critical awareness of practical and This will be taught throughout the
theoretical contexts in which data programme
scientists work
A5 Critically analyse the application of This will be taught throughout the
technology to real world problems programme
particularly in industry and
interdisciplinary research

Cognitive and thinking skills

Code Learning outcome Taught by the following module(s)

B1 Apply advanced skills and research-lead This will be taught throughout the
specialist knowledge in the design of programme
software and data analyses; academic
writing and presentation skills
B2 Critically analyse the application of This will be taught throughout the
technology to real world problems programme
particularly in industry and
interdisciplinary research

MSc Data Science - Programme Specification 4

Goldsmiths, University of London
Subject specific skills and professional behaviours and attitudes

Code Learning outcome Taught by the following module(s)

C1 Apply advanced skills and research-lead This will be taught throughout the
specialist knowledge in the design of programme
software and data analyses; academic
writing and presentation skills
C2 Critically analyse the application of This will be taught throughout the
technology to real world problems programme
particularly in industry and
interdisciplinary research

Transferable skills

Code Learning outcome Taught by the following module(s)

D1 Be able to do academic research and Data Science Research Topics, Final
writing, and present themselves and their Project
work
D2 Be able to reflect on and evaluate their This will be taught throughout the
work programme
D3 Be able to work effectively in groups This will be taught in group projects
which will be part of the assessment
in modules including Machine
Learning and Statistical Data Mining,
and Big Data Applications
D4 Be able to work effectively in groups All taught modules

Additional learning outcomes for the MSc:

Knowledge and understanding

Code Learning outcome Taught by the following module(s)

A1 Applied a deep understanding of cutting Final Project
edge technologies in the creation of a
substantial commercially and/or research-
wise relevant project

Cognitive and thinking skills

Code Learning outcome Taught by the following module(s)

B1 Propose, plan, execute and evaluate a Final Project
significant piece of original work

MSc Data Science - Programme Specification 5

Goldsmiths, University of London
Subject specific skills and professional behaviours and attitudes

Code Learning outcome Taught by the following module(s)

C1 Independently and in cooperation Final Project, but also the other
research and apply state of the art modules in various degrees, in
technologies in the context of concrete particular Machine Learning and
problems related to scalable data Statistical Data Mining, and Big Data
management and analytics Applications
C2 Design and program advanced computer Final Project, Machine Learning and
software and data products Statistical Data Mining, and Big Data
Applications, Introduction to Natural
Language Processing

Transferable skills

Code Learning outcome Taught by the following module(s)

D1 Be able to do academic research and Data Science Research Topics, Final
writing, and present themselves and their Project
work
D2 Be able to reflect on and evaluate their This will be taught throughout the
work programme
D3 Be able to work effectively in groups This will be taught in group projects
which will be part of the assessment
in modules including Machine
Learning and Statistical Data Mining,
and Big Data Applications
D4 Be proactive, plan their activity in This will be taught in throughout the
advance, and exercise personal programme
responsibility in their work

How you will learn

The Department of Computing is committed to a diverse and stimulating range of learning
and teaching methods that ensure the programme outcomes are addressed rigorously and
effectively. Learning emphasises a close synthesis between theoretical understanding and
practical application that helps students develop an advanced, critical approach to the
subject of Computing in general and to Data Science in particular.

The various modules of the programme provide a diverse range of topics. These will be
further developed through students’ independent research and learning activities directed
towards module assignments and the large-scale project component. The department is

MSc Data Science - Programme Specification 6

Goldsmiths, University of London
committed to providing a diverse and innovative range of teaching styles across its degree
programmes. These include traditional lecture and laboratory sessions but also a range of
more interactive and self directed activities focusing on independent, creative work and self
presentation. The nature of the learning activities will vary between different modules, but
includes data analysis, design and implementation (concerning several sorts of applications
in industry and/or interdisciplinary research), software design and coding, devising data
products, project planning, group activity and creative and research work.

Students are expected to engage in considerable independent reading and practical work
for all modules culminating in the final year project. This independent work will be supported
by library resources, access to lab space and computing cluster facilities, and supervision
from teaching staff.

How you will be assessed

The Department is committed to providing diverse types of assessment. Our methods of
assessment are designed to reflect research and business relevant activities and to
encourage independent and collaborative work. In particular our assessment integrates
different kinds of hands-on practical work including software and data products
development, data mining/analysis design and implementation, business planning, and
individual or group work. Students will be required to present their work in a number of
different ways including posters, traditional reports and short management reports, oral
presentations, and software and data products live or recorded presentations.

Each module in the programme will have its own type of assignment which corresponds to
the nature of the module. In addition to usual assignments, students will have a major final
project in the summer term, which should integrate what students have learnt throughout
the programme. The final project is an opportunity for students to work independently in a
large project reflecting the state of the art technology at a research level.

In collaborative assignment work particular care must be taken by students to describe

clearly and in detail precisely the nature of their contributions, and the contributions of their
group collaborators. This must be delivered as part of students' assignment written reports
and/or evaluation documentation, as required in the documentation brief. In particular
students should report on the group dynamics, explain their role in the group, and indicate
any problems that the group encountered and how they have been addressed.

Feedback is very important to the learning process, and shows students how to improve
their work, and provides suggestions on how to learn more effectively in the future.
Therefore the Department is committed to providing timely and full feedback on all assessed
assignments.

MSc Data Science - Programme Specification 7

Goldsmiths, University of London
Final projects will be assessed based on the submission of a final report and a presentation
in a viva. Guidance on the structure and writing of the report will be given in the module
handbook. Moreover, general guidance on writing scientific work will specifically be provided
in the Data Science Research Topics module. Projects will be marked by a panel composed
of two members of academic staff.

Students who are unable to submit an assessment on time due to illness or other
unavoidable circumstances, must provide documentary evidence to their personal tutor in
order to be allowed a late submission. Evidence must also be supplied for students to apply
for consideration of mitigating circumstances in assessment.

Marking criteria

Mark Descriptor Specific Marking Criteria

80-100% Distinction A grade in the range of 80-100% will be awarded in the case of
(Outstanding/ really accomplished work that demonstrates high levels of
Exceptional) scholarship and originality. This grade will reflect the overall
achievement of the appropriate learning outcomes to an
exceptionally accomplished level. In particular a grade in the
90s should be reserved for work deemed to be outstanding,
and of publishable quality.
70-79% Distinction A grade in the range of 70-79% will be awarded when
candidates show evidence of an excellent application of
appropriate knowledge, understanding and skills as specified in
the module learning outcomes. Demonstration of a thorough
grasp of relevant concepts, methodology and content
appropriate to the subject discipline; indication of originality in
application of ideas, in synthesis of material or in performance;
insight reflects depth and confidence of understanding of the
material.
60-69% Merit Demonstration of a deep level of understanding based on a
competent grasp of relevant concepts, methodology and
content; display of skill in applying interpreting complex
material; organisation of material at a high level of
competence. Students should be able to demonstrate the
ability to work independently to research and implement state
of the art technologies.
50-59% Pass Demonstration of a sound level of understanding based on a
competent grasp of relevant concepts, methodology and
content; display of skill in organising, discussing and applying
complex material. Students should be able to implement state
of the art technologies under guidance.

MSc Data Science - Programme Specification 8

Goldsmiths, University of London
 Mark Descriptor Specific Marking Criteria
30-49% Fail Represents an overall failure to achieve the appropriate
learning outcomes. Students achieve some of the aims but
were unable to demonstrate independence and originality
beyond what would be expected at undergraduate level.
10-29% Bad fail Represents a significant overall failure to achieve the
appropriate learning outcomes.
1-9% Very bad fail A submission that does not attempt to address the modules
specified learning outcomes. It will be considered a non-valid
attempt and the module must be re-sat.
0% Non Work was not submitted, or it was plagiarised.
submission or
plagiarised

How the programme is structured

MSc Data Science consists of two terms of taught modules followed by a large-scale
project. The taught modules introduce you to fundamental mathematical and computational
skills and show you how to apply them to real world data.

The programme includes:

• A firm grounding in the theory of data mining, statistics and machine learning
• Hands-on practical real world applications such as social media, biomedical data and
financial data with Hadoop (used by Yahoo!, Facebook, Google, Twitter, LinkedIn,
IBM, Amazon, and many others), R and other specialised software
• The opportunity to work with real-world software such as Apache

In additional to the compulsory programme modules, students choose 3 optional modules

from the indicative list below.

Full-time mode

Module Title Module Code Credits Level Module Status Term

Machine Learning and IS71060A 30 7 Compulsory 1,2
Statistical Data Mining
Big Data Applications IS71059B 15 7 Compulsory 2
Data Science Research IS71058A 15 7 Compulsory 2
Topics
Data Programming IS71068A 15 7 Compulsory 1

MSc Data Science - Programme Specification 9

Goldsmiths, University of London
 Module Title Module Code Credits Level Module Status Term
Three 15 credit optional Various 45 7 Optional 1 and
modules totalling 45 credits 2
from the list below:
Neural Networks IS71040A 15 7 Optional 1
Open Data IS71063A 15 7 Optional 1
Data Visualisation and the IS71082A 15 7 Optional
Web
Data Compression IS71067A 15 7 Optional 1
Multimedia Informatics I: IS71064A 15 7 Optional 2
Music Information Retrieval
and digital musicology
Geometric Data Analysis IS71069B 15 7 Optional 2
Final Project IS71061A 60 7 Compulsory 3

Part-time mode

Academic year of study 1

Module Title Module Code Credits Level Module Status Term

Machine Learning and IS71060A 30 7 Compulsory 1,2
Statistical Data Mining
Big Data Applications IS71059B 15 7 Compulsory 2
Data Programming IS71068A 15 7 Compulsory 1

Academic year of study 2

Module Title Module Code Credits Level Module Status Term

Final Project IS71061A 60 7 Compulsory 3
Data Science Research IS71058A 15 7 Compulsory 2
Topics
PLUS 3 x 15 CAT optional Various 45 7 Compulsory 1 or 2
modules from the full-time
mode above

Academic support
Support for learning and wellbeing is provided in a number of ways by departments and
College support services who work collaboratively to ensure students get the right help to
reach their best potential both academically and personally.

MSc Data Science - Programme Specification 10

Goldsmiths, University of London
All students are allocated a Personal Tutor (one in each department for joint programmes)
who has overall responsibility for their individual progress and welfare. Personal Tutors
meet with their student at least twice a year either face-to-face, as part of a group and/or
electronically. The first meeting normally takes place within the first few weeks of the
autumn term. Personal Tutors are also available to students throughout the year of study.
These meetings aim to discuss progress on modules, discussion of the academic discipline
and reports from previous years if available (for continuing students). This provides an
opportunity for progress, attendance and assessment marks to be reviewed and an
informed discussion to take place about how to strengthen individual learning and success.

All students are also allocated a Senior Tutor to enable them to speak to an experienced
academic member of staff about any issues which are negatively impacting their academic
study and which are beyond the normal scope of issues handled by Programme Convenors
and Personal Tutors.

Students are provided with information about learning resources, the Library and
information available on Learn.gold (VLE) so that they have access to department/
programme handbooks, programme information and support related information and
guidance.

Taught sessions and lectures provide overviews of themes, which students are encouraged
to complement with intensive reading for presentation and discussion with peers at
seminars. Assessments build on lectures and seminars so students are expected to attend
all taught sessions to build knowledge and their own understanding of their chosen
discipline.

All assessed work is accompanied by some form of feedback to ensure that students’ work
is on the right track. It may come in a variety of forms ranging from written comments on a
marked essay to oral and written feedback on developing projects and practice as they
attend workshops.

Students may be referred to specialist student services by department staff or they may
access support services independently. Information about support services is provided on
the Goldsmiths website and for new students through new starter information and
induction/Welcome Week. Any support recommendations that are made are agreed with the
student and communicated to the department so that adjustments to learning and teaching
are able to be implemented at a department level and students can be reassured that
arrangements are in place. Opportunities are provided for students to review their support
arrangements should their circumstances change. The Disability and Wellbeing Services
maintain caseloads of students and provide on-going support.

MSc Data Science - Programme Specification 11

Goldsmiths, University of London
The Careers Service provides central support for skills enhancement, running The Gold
Award scheme and other co-curricular activities that are accredited via the Higher Education
Achievement Report (HEAR).

The Academic Skills Centre works with academic departments offering bespoke academic
literacy sessions. It also provides a programme of academic skills workshops and one-to-
one provision for students throughout the year.

Links with employers, placement opportunities and career

prospects
The MSc Data Science develops analytic and critical skills, providing successful students
with the tools and competencies needed to intelligently interrogate numerical, textual and
qualitative data; to extract meaning from raw information; and to communicate the results of
their investigations, and their implications, to stakeholders or other interested parties. These
skills lead naturally to embarking on a variety of careers, with employers from the financial
sector, technology firms small and large, biomedical research sector, the charitable and
voluntary sector, and public research sector. The mix of attributes encouraged by the
programme as technical ability to manage and process data, reflection and insight to
develop understanding, and empathy and awareness to communicate it effectively is highly
desirable to prospective employers. The programme's structure, in particular around the
final project and preparation for it, encourages student engagement with projects requiring
the services of data scientists, and will provide networking opportunities to start the students
along their chosen career path. Our graduates will be exposed to the ethical issues of data
science; in the modern era of data availability, it is crucial that all participants in data
exchange and treatment are aware of the impact of their behaviour on privacy, anonymity
and personal security.

The requirements of a Goldsmiths degree

All taught postgraduate degrees have a minimum total value of 180 credits and involve one
calendar year of full-time study. Some programmes may extend over more than one
calendar year and, when this is the case, they have a higher total credit value. Programmes
are composed of individual modules, each of which has its own credit value. Part-time
students normally take modules to the value of 90 credits each year. If a programme has a
part-time pathway, the structure will be set out in the section “How the programme is
structured” above. Normally, all modules are at level 7 of the Framework for Higher
Education Qualifications.

More detailed information about the structure and requirements of a Goldsmiths degree is
provided in the Goldsmiths Qualifications and Credit Framework.

MSc Data Science - Programme Specification 12

Goldsmiths, University of London
Modules

Modules are defined as:

• “Optional” – which can be chosen from a group of modules

• “Compulsory” – which must be taken as part of the degree

Progression

Some programmes may require students to pass specific modules prior to completion of the
dissertation/major project (or equivalent). Additionally, where a programme of study extends
beyond one calendar year, students may be required to pass specific modules in their first
year of study before progressing to the second year. Where this is the case, these
requirements will be set out in this Programme Specification.

Award of the degree

In order to graduate, students must successfully complete all modules specified for the
programme, as set out within the section “How the programme is structured” above.

Classification

Final degree classification is calculated on the basis of a student’s mean average mark
(based on credit value) across all modules on the programme.

Masters degrees are awarded with the following classifications:

Distinction – 70%+
Merit – 60-69%
Pass – 50-59%

More detail on the calculation of the final classification is on our website.

Interim exit awards

Some programmes incorporate interim exit points of Postgraduate Certificate and/or

Postgraduate Diploma, which may be awarded on the successful completion of modules to
the minimum value of 60 credits or 120 credits respectively. The awards are made without
classification.

MSc Data Science - Programme Specification 13

Goldsmiths, University of London
When these awards are incorporated within the programme, the relevant learning outcomes
and module requirements will be set out within the “What you will be expected to achieve”
section above.

The above information is intended as a guide, with more detailed information available in
the Goldsmiths Academic Manual.

Programme-specific rules and facts

In order to progress to the Final Project, students must fulfil the requirements for a pass at
Postgraduate Diploma level (pass all 120 credits of taught modules).

General programme costs

In addition to your tuition fees, you will be responsible for meeting standard costs
associated with your study. Find out more information at gold.ac.uk/programme-costs.

Specific programme costs

Not applicable.

How teaching quality will be monitored

Goldsmiths employs a number of methods to ensure and enhance the quality of learning
and teaching on its programmes.

Programmes and modules are formally approved against national standards and are
monitored throughout the year, such as in departmental committees, a variety of student
feedback mechanisms and through the completion of module evaluation questionnaires.
Every programme has at least one External Examiner who reviews comments annually on
the standards of awards and student achievement. External Examiner(s) attend Boards of
Examiners meetings and submit an annual written report.

Modules, programmes and/or departments are also subject to annual and periodic review
internally, as well as periodic external scrutiny.

Quality assurance processes aim to ensure Goldsmiths’ academic provision remains

current, that the procedures to maintain the standards of the awards are working effectively
and the quality of the learning opportunities and information provided to students and
applicants is appropriate.

Detailed information on all these procedures are published on the Quality Office web pages.

MSc Data Science - Programme Specification 14

Goldsmiths, University of London