Note: Some of the graphical elements of this site are only visible to browsers that support accepted web standards. The content of this site is, however, accessible to any browser or Internet device.
A guide to Individual Projects for MSc in Computing for Industry
Introduction
The individual project is in many ways the most important single component of the MSc programme. It provides the opportunity for you to demonstrate independence and originality, to plan and organise a large project over a long period, and to put into practice the techniques you have been taught throughout the course. Whatever your level of academic achievement so far, you can show your individuality and inspiration in this project. It should be the most satisfying piece of work in your course.
The Project Coordinator
This year the Project Coordinators are James Jacobson jj@doc.ic.ac.uk & Connie Bao hb1@doc.ic.ac.uk and are responsible for the general organisation of MSc individual projects.
Choosing an Individual Project
We expect that in most cases your project will be carried out at your workplace and be concerned with a topic of interest and relevance to your professional activities. The idea for your project may either come from you or from your employer, or it may be a proposal from a member of staff or perhaps a combination of the two. If you see a project on the existing projects list that interests you, then contact the supervisor for further information. In any case you will have a project supervisor from the Department and usually also a mentor at your workplace. Projects that are proposed by your employer or yourself will need to be associated with a staff member before being approved. You are welcome to discuss any ideas with the co-ordinator, who will look for a suitable internal supervisor (although you cannot assume one will be found in every case). Projects on the main project proposal list already have an internal supervisor and you are encouraged to find a supervisor from your workplace as well.
Some projects may be of a sensitive topic, in which case the internal supervisor will sign a suitable declaration of confidentiality lasting for a period of up to 2 years.
Once you have finalised your project with your supervisor, you need to complete a registration form which can be downloaded from here.
The form requires your details and needs the signatures of your supervisor and normally of a representative of your employer. When completed the form should be sent to Barbara Claxton.
Assessment
General requirements. All projects should include some element of original work. An implementation project could develop a new application, or enhance some existing application by improving its performance of functionality. Appropriate implementation details, such as source code, diagrams, etc, should normally be included in the Appendix. Projects which are predominantly survey reports must be backed up by experimentation, implementation, analysis, new examples, and so on. Your supervisor will advise on how to develop your project appropriately.
A project normally takes the equivalent of four months full time to complete, although you have a maximum of 1 year (from the time your project is finally arranged) in which to submit. Remember that a good project involves a lot of input from you. A distinction level project involves a combination of sound background research, a solid implementation, or piece of theoretical work, and a well-structured and well presented report detailing the project's background, objectives and achievements. The project will involve significant technical problems that have been overcome.
Assessment. All (successful) MSc dissertations are lodged in the College Library and are open to public scrutiny. (There are exceptions to this rule for projects that involve sensitive material, in which case it can be arranged that the project be withheld for up to 2 years.) Your project work is assessed on the basis of your dissertation and the quality and substance of the work done. However, you cannot compensate for a poor write-up by a well-conducted project; and you cannot get credit for ideas or experiments not included in the dissertation. Dissertations not meeting minimal standards of presentation will not be accepted for award of the MSc degree, no matter how good the project work itself.
Practicalities
You will probably use your own (or workplace) equipment on which to develop software or hardware. In case you need additional software which is not currently provided by the Computing Support Group (CSG) you must complete a Resource Request Form (available from the Technical Library room 219). A request can then be made to purchase it if an acceptable alternative is not available. A purchase request will need the support of your supervisor and is not guaranteed to be approved.
Please note that there is no excuse for failing to keep adequate backups and if you lose your program or your data or your dissertation because of a system failure no allowance can be made. Extensions will not be given at the end of the project for you to re-type a lost dissertation, for example.
Meeting Your Supervisor
You must make sure that you arrange regular meetings with your supervisor. The meetings may be brief once your project is under way but your supervisor needs to know that your work is progressing. You should make sure your supervisor can contact you (e.g. give them your address) if necessary. If you need to talk to your supervisor between scheduled meetings use electronic mail to communicate your problem. Prepare a written list of points you wish to discuss before you go to see your supervisor (or second marker). Take notes during the meeting so that you do not forget the advice you were given or the conclusions that were reached.
The Dissertation
The dissertation is an extremely important part of the project. It serves to show what you have achieved and should demonstrate that:
You understand the wider context of computing by relating your choice of project, and the approach you take, to existing products or research. You can apply the theoretical and practical techniques taught in the course to the problem you are addressing, and that you understand their relevance to the wider world of computing. You are capable of criticising your own work objectively and making constructive suggestions for improvements or further work based on your experiences so far. You can explain your thinking and working processes clearly and concisely to third parties who may not be experts in the field in which you are working.
Remember that second markers, and other readers, will not have followed the project throughout. Make the presentation reasonably self-contained. State the objectives clearly; provide sufficient background material.
Many students underestimate the importance of the dissertation. You should consider that the aim of the project is to produce a good dissertation and that software, hardware, theory etc. that you develop during the project are merely a means to this end. Do not make the mistake of leaving the write-up to the last minute. Ideally you should produce the bulk of the dissertation as you go along and use the last few weeks to bring it together into a coherent document.
The physical layout and formatting of the report is also important, and yet is very often neglected. A tidy, well laid out and consistently formatted document makes for easier reading and is suggestive of a careful and professional attitude towards its preparation.
Remember that quantity does not automatically guarantee quality. A 150 page dissertation is not twice as good as a 75-page one, nor a 10,000 line implementation twice as good as a 5,000 line one. Conciseness, clarity and elegance are invaluable qualities in report writing, just as they are in programming, and will be rewarded appropriately. Also, it is important to appreciate that the appropriate size and structure of a dissertation can vary significantly from one project to the next. As a guide, most project reports are between 50 and 100 pages in length. Despite these variations, however, most good dissertations have the following components in common:
Presentation
The dissertation must be bound in the University approved manner (a list of binders who can do this is given in the Appendix), although it is likely that in hte near future a soft bound copy will be acceptable.
Title page: This has the standard form given in the Appendix.
Abstract: The abstract is a very brief summary of the dissertation's contents. It should be about half a page long. Somebody unfamiliar with your project should have a good idea of what it is about having read the abstract alone and will know whether it will be of interest to them.
Acknowledgements
It is usual to thank those individuals who have provided particularly useful assistance, technical or otherwise, during your project. Your supervisor will obviously be pleased to be acknowledged as he or she will have invested quite a lot of time overseeing your progress.
Contents page
This should list the main chapters and (sub) sections of your dissertation. Choose self-explanatory chapter and section and if possible include page numbers indicating where each chapter/section begins. Try to avoid too many levels of subheading. Try if possible to stick to sections and subsections; sub subsections are usually avoidable.
Introduction
This is one of the most important components of the dissertation. It should begin with a clear statement of what the project is about so that the nature and scope of the project can be understood by the reader. It should summarise everything you set out to achieve, provide a clear summary of the project's background and relevance to other work and give pointers to the remaining sections of the dissertation which contain the bulk of the technical material.
Background
The background section of the dissertation should set the project into context by relating it to existing published work (or unpublished work on which the project builds). The background section is sometimes included as part of the introduction but more usually is a separate chapter, or collection of chapters if the project involved an extensive amount of research. The published work may be in the form of research papers, articles, text books, technical manuals, or even existing software or hardware of which you have had experience. You must acknowledge the sources of your inspiration; you are expected to have seen and thought about other peopleís ideas; your contribution will be putting them into practice or developing them in some new direction. One rule is clear: if you present another person's work as your own and do not cite your sources of information/inspiration you are cheating. When referring to other pieces of work, cite the sources at the point they are referred to or used, rather than just listing them at the end. The University of London takes a very strict line on plagiarism, and its standard notice on the subject is given in the Appendix. We are required to draw it to your attention.
Body of report
The central part of the report usually consists of three of four chapters detailing the technical work undertaken during the project. The structure of these chapters is highly project dependent. Usually they reflect the chronological development of the project, e.g. design, implementation, experimentation, optimisation, although this is not always the best approach. However you choose to structure this part of the report, you should make it clear how you arrived at your chosen approach in preference to the other alternatives documented in the background. For implementation projects you should describe and justify the design of your program at some high level, for example by using any of the design methods taught during the first and second term courses, and should document any interesting problems with, or features of, your implementation. Integration and testing are also important to describe. Your supervisor will advise you on the most suitable structure for these middle sections.
Conclusions and Future Work
All good projects conclude with an objective evaluation of the project's successes and failures and suggestions for future work which can take the project further. It is important to understand that there is no such thing as a perfect project. Even the very best pieces of work have their limitations and you are expected to provide a proper critical appraisal of what you have done. Your assessors are bound to spot the limitations of your work and you are expected to be able to do the same.
Bibliography
This consists of a list of all the books, articles, manuals etc. used in the project and referred to in the report. You should provide enough information to allow the reader to find the source. You should give the full title and author and should state where it is published, including full issue number and date, and page numbers where necessary. In the case of a text book you should quote the name of the publisher as well as the author(s).
Appendix
The appendices contain information which is peripheral to the main body of the report. Information typically included are things like program listings, tables, proofs, graphs or any other material which would break up the theme of the text if it appeared in situ. Large program listings are rarely required, and should be compressed as much as possible, e.g. by printing in multiple columns and by using small font sizes, omitting inessential code etc.
User Guide
For projects which result in a new piece of software you should provide a proper user guide providing easily understood instructions on how to use it. A particularly useful approach is to treat the user guide as a walk-through of a typical session, or set of sessions, which collectively display all the features of your package. Technical details of how the package works are rarely required. Keep it concise and simple. The extensive use of diagrams illustrating the package in action prove particularly helpful. The user guide is sometimes included as a chapter in the main body of the report, but is often better as an appendix to the main report. Do not include User Guides for trivial pieces of code where these are not the main subject of the dissertation.
Assessment of the Dissertation
The assessment of the dissertation will be undertaken by the internal supervisor and a second marker in the first instance. A selection of dissertations, and all dissertations which are candidates for distinction or borderline, are assessed by several other persons, and by an external examiner.
Appropriate implementation details, such as source code or circuit diagrams, should normally be included as an appendix, together with indications of sample runs. Projects which are predominantly survey reports must be backed up with experimentation, implementation, theoretical or conceptual analysis, new illustrative examples, and so on.
A distinction level project is one which demonstrates significant breadth and depth. It involves a combination of sound background research, an outstanding implementation or an outstanding piece of theoretical work, and a well-structured and well-presented report detailing the project's background, objectives and achievements. The project involves significant technical problems that have been largely, or completely, overcome. Dissertations should be assessed by the following broad criteria:
Background Preparation
This assesses the way you arrived at your initial project specification, work programme and list of objectives. It particularly addresses the background research undertaken and the manner in which your approach and programme of work fits in with the current state-of-the-art. It assesses your awareness of relevant previous work, and how the project builds upon or exploits existing techniques or results.
Technical Achievement : this assesses the main technical output from the project. It addresses specific issues such as the design, correctness, elegance, usability etc. of the final product, theoretical or practical, the approach and techniques employed, and the significance of the achievements relative to the state-of-the-art.
Quality of Dissertation : selection, organisation and presentation of material, quality of prose, clarity of explanations, discussion of related work, spelling, punctuation, legibility, clarity and relevance of diagrams and so on. Note that reports falling below minimum acceptable standards may not be accepted no matter how good the rest of the project work is.
Remember that one of your supervisor's main responsibilities is to advise on how to write-up your project results. You are not expected to be able to produce a perfect dissertation without help. You should discuss with your supervisor all aspects of the dissertation, but particularly its structure and how to present the material.
You might find it useful to look at MSc dissertations from previous years. They are all available in the College Library. Your supervisor may also have examples.
Grading and Minimum Requirements The pass mark for a project is 50% and the Distinction mark is 70%.
It is usual to arrange a demonstration (and/or presentation) of your project for your supervisor and second marker (and anyone else who may be interested in the project, for example your mentor at your place of work).
Submitting Your Dissertation
- You are required to hand in two copies of your dissertation and electronically submit your dissertation. (Currently these should be bound, but please check as soft copies may also be acceptable.)
- Dissertations must be submitted to the Student Admin Office (Room 345) and electronic copies should be sent to Connie Bao. Before your dissertation will be accepted you must:
- Sign and submit the receipt for electronic submission which will be sent to you when you provide the electronic copy of your dissertation.
- If the project is your last module for the course you must provide a clearance form from the Lyon Playfair Library which will be given to you by the librarian when you have returned all your library books; note that this clearance also refers to the Departmental Technical Library.
Pitfalls
- Some of the most useful things to know about individual projects are the common pitfalls. Why do some projects go wrong? Here are some of the common causes of failure:
- Failing to meet your supervisor regularly. If you arrange a meeting with your supervisor, turn up at the agreed time. If you are stuck for any reason and you have no meeting arranged, contact him or her immediately. You gain no sympathy from anyone if you lose contact with your supervisor and produce a poor project as a result. Your supervisor will be happy to help you but they can do nothing if they are unaware that your are having trouble.
- Allowing too little time for the report. You should try to produce as much of your report as you can as you go along, even though you don't know in advance its exact structure. The last few weeks of the project should be dedicated to pulling together the material you have accumulated and producing a polished final product.
- Failing to plan a fall-back position if the planned work is not completed on time. Try to plan your project in stages so that if things go wrong in a later stage you have a completed stage to fall back on.
- Over/Under Ambition. Try to be realistic about what you can achieve in the time available. A good project requires a lot of input from you and should prove to be technically challenging throughout. At the same time, however, it is better to do a small job well than it is to fail to do a big job at all. Your supervisor will advise you on his or her expectations of the project and this will help you set your sights accordingly.
- Submission of preliminary drafts. Do not submit your thesis without letting your supervisor read through it first. He or she will invariably have comments and suggestions for improvement. You should normally expect to revise the complete draft at least once. Dissertations failing to meet minimum standards will not be accepted for award of the MSc degree.
APPENDIX
Typing and Binding
The dissertation should be typed on A4 paper with an adequate binding margin (about 4 cm) on the left hand side. Two copies are required, both of which are retained by the College. (In case of issues of confidentiality, the Department will sign a declaration to withhold the project from any public scrutiny for up to 2 years.) The title page of the dissertation should take the following standard form:
Imperial College London
Department of Computing
Title of Dissertation
by
Name and Initials
Your attention is also drawn to the following notice issued by the University of London
You are reminded that all work submitted as part of the requirements for any examination of the University of London must be expressed in your own words and incorporate your own ideas and judgements. Plagiarism: that is, the presentation of another person's thoughts or words as though they were your own must be avoided, with particular care in course-work and essays and reports written in your own time. Direct quotations from the published or unpublished work of others must always be identified as such by being placed in quotation marks, and a full reference to their source must be provided in the proper form. Remember that a series of short quotations from several different sources, if not identified as such, constitutes plagiarism just as much as does a single unacknowledged long quotation from a single source. Equally, if you summarise another person's ideas or judgements, you must refer to that person in your text, and include the work referred to in your bibliography. Failure to observe these rules may result in an allegation of cheating. You should therefore consult your tutor or course director if you are in any doubt about what is permissible.