MSc Computational Intelligence and Computational Biology

Entry standards

Candidates should have a first degree in computer science, mathematics, electrical engineering or physics. Students from neuroscience, bioscience and other related disciplines having an adequate mathematical and computing background are also encouraged to apply. Candidates should have obtained the degree at Upper Second level or higher. In exceptional circumstances, relevant work experience may also be considered if the candidate has achieved less than Upper Second.

English language requirements

Non-native speakers of English will normally be required to have IELTS 6.5 or above, with a minimum of 6.0 in each component (or equivalent).). Please note that the University of Surrey offers English language programmes and is also an IELTS Test Centre.

Please note that the University of Surrey offers English language programmes and is also an IELTS Test Centre.

MSc Computational Intelligence and Computational Biology - structure and modules

Compulsory Modules

Computational and Cognitive Neurosciences

This module introduces the backgrounds of computational and cognitive neurosciences, including spiking models of neurons and neuronal networks, neurodynamics, learning algorithms such as Hebbian learning and spiking timing dependent plasticity. In addition, the main components of the brain involved in memory and cognition and behaviour control will also be described.

Introduction to Computational Systems Biology

This module describes the fundamental relationships and interactions between various parts of a biological object by which it is hoped that an understandable model of the whole system can be developed. The module will cover chemical foundation and enzymes, and fundamentals of molecular biology focusing on a systems level, by modelling the gene expression process, that is. the gene regulatory networks, methodologies for gene regulatory network, kinetic model of gene circuit, signal transduction, and dynamic signalling and gene expression regulation.

Bioinformatics and Bioinformatics Programming

This module provides knowledge of the current bioinformatic analytical and prediction techniques for the analysis and interpretation of large-scale ‘-omics’ data sets central to the field of systems biology. Methods to be described include: genomics, for identifying genetic differences between species via sequence analysis and comparative genomic hybridisations (CGH); transcriptomics and chromatin immunoprecipitation on a chip (ChIP-chip), for the construction of putative gene regulatory networks via gene expression profiles, regulatory protein enrichment, clustering and promoter predictions; proteomics, for calculating protein abundance measurements and protein structure. An introduction to the statistical programming environment R and the programming language Perl will be included. In addition, a grounding in the principles of molecular biology, the founding biological science behind many high-throughput quantitative ‘-omics’ technologies, will be given.

Brain–Computer Interface

Brain–machine interface is a rapidly developing research area that greatly helps understanding the functions of the brain from generation of active potentials to modelling of various states of the brain. This module introduces the basic sensor technologies for capturing brain signals, the concepts of brain motor function, synchronisation and de-synchronisation of the brain, connectivity of brain regions, and the brain response to various stimuli (event related and evoked potentials) by means of signal processing theory and methods. In addition, implications of neuro-feedback will be discussed and examined.

Evolutionary Computation and Artificial Development

Evolution and development are two areas where a systems approach can play an important role. On the basis of basic knowledge in systems biology, this module describes computational models for studying biological evolution and development.
Starting from a standard direct-coded evolutionary system, the module will discuss in detail how evolutionary developmental systems can be built and investigated in a computing environment.
Then models for neural and morphological development will be presented. Possible engineering applications of such systems will also be discussed.

Optional modules include:

• Collective Intelligence
• Database and Knowledge Discovery
• Cloud Computing

Subject information

Our Degree Programmes

The Department offers four MSc programmes. Each with its own distinct focus, all of them offer a taste of the Department’s specialisms.

Internet Computing, Security Technologies and Applications, and Computational Intelligence and Computational Biology are technical degrees requiring a solid background in computing or a cognate discipline.

The MSc in Information Systems is more of a generalist degree, accepting students from a wider range of backgrounds. It takes a more high-level, overall view on information technologies and prepares students for managerial roles rather than technical roles in their future careers. It covers business and management topics as well as technical computing subjects.

The MSc in Security Technologies and Applications is specialised towards the Department’s multiple activities in relation to the technologies and principles that underlie a variety of information security techniques and technologies.

The MSc in Internet Computing is concerned with distributed information and computing resources, and builds from the Department’s activities relating to the Web and the Cloud. The programme also has a strong element of more general, technical computer science and software development skills.

The MSc in Computational Intelligence and Computational Biology is the latest addition to the portfolio of programmes, following the appointment of a Professor of Computational Intelligence and the subsequent growth of a nature-inspired computing theme in the Department.

All of the programmes benefit from the strong research community and industrial partnerships of the Department. In particular, the dissertation project allows students to work on a topic in one of the key research areas. Even though the programmes share many modules, each has its own focus and direction.

Regardless of which programme you choose, we are committed to making your year at Surrey a valuable and enjoyable experience.

MSc Computational Intelligence and Computational Biology - entry standards

Entry standards

English language requirements

Non-native speakers of English will normally be required to have IELTS 6.5 or above, with a minimum of 6.0 in each component (or equivalent).). Please note that the University of Surrey offers English language programmes and is also an IELTS Test Centre.

Please note that the University of Surrey offers English language programmes and is also an IELTS Test Centre.

Start date

September

Programme Director

• Professor Yaochu Jin

MSc Computational Intelligence and Computational Biology - fees and funding

Fees

Computational Intelligence and Computational Biology (full time):

UK/EU - £6,400 
Overseas - £14,440

www.surrey.ac.uk/pgfees/2012

MSc Computational Intelligence and Computational Biology - professional context

Industry Sponsors and Prizes

The Department benefits greatly from strong links with industry, and our industry partners support the programme in various ways; some with guest lectures and some with prizes for the best student performance (typically £150–200 per prize).

Scholarships

The Department pays a scholarship of £1000 to students with a First class honours degree or equivalent, who do not receive any other scholarship or bursary from the University. Please see the Department web pages for the full conditions.

MSc Computational Intelligence and Computational Biology - teaching

Teaching

Taught Masters programmes in the Department of Computing utilise our research-active staff in conjunction with state-of-the-art facilities. We provide a range of learning experiences including lectures, tutorials, directed study, practical laboratories and project work that prepare graduates for their professional life.

We are particularly keen to develop, in all our students, a broad range of generic skills to complement the core technical or scientific competencies of their chosen subject area. Our modular programme format, coupled with the increasing use of innovative teaching and learning strategies such as e-learning and industrially focused short courses, provides a flexible study environment whilst maintaining academic rigour and quality. Our record of graduate employment is outstanding, with Surrey graduates consistently being in high demand across all sectors.

MSc Computational Intelligence and Computational Biology - learning

Dissertation and Projects

Dissertation Project

The MSc dissertation project makes up one third of the degree programme, starting at the end of the first semester and completing at the end of the summer. During the project, you are supervised by a member of academic staff to advise and guide you to completion. At the end of the project you must submit your bound dissertation, which forms a complete record of the project, which is then held in the University Library.

The project focuses in depth on a subject at the leading edge of computing. For example, projects can undertake the development of a software system to solve a particular problem, possibly in collaboration with an industrial partner.

Alternatively, projects can be research-based, in which case an aspect of computing is investigated, perhaps to evaluate particular techniques or propose a new algorithm. These projects are usually closely linked to the Department’s research strengths.

Whatever the topic, you are expected to develop a critical understanding of the methods and technologies needed, then implement and evaluate your chosen solution to a professional standard. Project planning and time management is an important part of the experience.

Project Examples

David Lundin – E-voting Project 

Elections across the globe have become high-profile events, not least because of their controversy, such as the American presidential election of 2000. One way in which elections can be made more reliable is through the use of electronic voting systems, which can provide security and verifiability.

Some electronic voting systems, such as Prêt à Voter and Punchscan, use a pre-printed paper ballot form, part of which is destroyed to create an encrypted receipt of the vote. The voter can use this receipt to check online that his or her vote has been included in the tally, but as the receipt is encrypted, it cannot be used to prove which candidate the vote is for. The problem with these ballot forms is that anyone who can see them before they are used has sufficient knowledge to check the contents of an encrypted receipt without it having to be properly decrypted.

In David’s MSc project he developed a method based on visual encryption of the candidate list that hides the content of the ballot form until the moment when it is used by the voter in the voting booth. When the top layer, printed on the ballot form, is properly aligned over the bottom layer, displayed on a computer screen, the candidate list appears in plain text. When the form is removed from the screen, the content of the form is once again hidden.

Ian Golledge – Identifying and Classifying Electronic Spam

The project presents a prototype model for implementing a self-organising map as a spam filter. A method is shown where emails are converted into feature vectors, where features represent keywords. Keywords are selected from an analysis of an email corpus with the results ranked based on word frequency measurements. The project describes phases of design which attempt to improve on feature selection and conclude on a prototype model for spam filtering using the self-organising maps.

This prototype model is evaluated over six datasets of fluctuating ratios of ham and spam, with testing designed to emulate the incremental re-training of a personalised spam filter. The results are compared to common techniques in spam filtering. Initial results show the model can outperform popular Naive Bayesian techniques. The feature vector representation is then further developed and the model shows results that compare strongly against other classifiers identified in research, demonstrating effective application of selforganising maps for spam filtering.

The work was published at an international IEEEsponsored conference in Italy: Vrusias, B. & Golledge, I. (2008). Adaptable text filters and unsupervised neural classifiers for spam detection. Advances in Soft Computing, 53, 195–202.

Aaron Randall – Authentication and Self- Restoration of Watermarked Images

In an age where digital media use is prolific, accessible and cost-effective, the requirement for digital images to be used in such situations as evidence in court, medical imaging, traffic enforcement and forensics is increasingly important.

However, along with digital media use, many different techniques to alter media files digitally have been developed – some with very realistic results. The ramifications of using potentially tampered digital media as evidence in court, for example, could cause the difference between an innocent or guilty verdict. This is an issue that digital watermark authentication and restoration attempts to answer.

Aaron’s MSc project was to develop an image authentication and restoration system capable of localising manipulated regions of an image. Through iterative restoration techniques and the extraction of hidden data from the image itself, he would then attempt to restore damaged regions. The first step of authentication highlighted any regions of the image that the application believed to have been tampered with in some way (such as cropping, blurring or other image manipulation techniques). The restoration stage then looked at fixing the regions highlighted as tampered. This step extracted hidden data from the image and inserted it back into the relevant damaged regions.   

MSc Computational Intelligence and Computational Biology - graduate profile

MSc Computational Intelligence and Computational Biology - more

Our Department

The Department of Computing is an active department within the Faculty of Engineering and Physical Sciences. There are 17 full-time academic staff and around 400 students at all levels - from undergraduate through to PhD. The Department is proud of its friendly reputation and aims to provide a supportive environment for students.

The Computing degree programmes at the University of Surrey and the all-round student experience is highly rated in the National Student Survey and league table results. The undergraduate and MSc degree programmes are stimulating and challenging with a high level of practical content. The Department is constantly exploring the use of new technology in teaching - such as podcasts and voting handsets - to support greater interactivity, accessibility and enjoyment of lectures. They also organise extra-curricular activity for those who want to explore other aspects of computing in more depth, while supporting local schools and colleges to bring some of the exciting aspects of computing to school pupils.

The Department's research interests are many and varied - ranging across the research groups of Digital Ecosystems; Formal Methods and Security; Multimedia Security and Forensics; and Nature Inspired Computing and Engineering. With a strong research culture and a growing research profile, the Department of Computing have won several prizes - including the University Voting Systems Competition System Design award in 2007 and the prestigious Institution of Engineering and Technology Innovation in Engineering Security award in 2006.

MSc Computational Intelligence and Computational Biology - apply

• MSc Computational Intelligence and Computational Biology - starting September 2012