Graham Kendall

3rd Year Undergraduate Project Dissertation

Introduction

This page gives some advice as to how third year students should structure their dissertations. It is still in your interests to let me see the dissertation before you hand it in - this page is just meant to give you general advice which I find myself repeating to many of my students. There is advice on both the interim dissertation as well as the final dissertation. In addition, you might like to consider writing your dissertatin into a format that could be submitted to a conference or a journal. This could gretaly increase your chances of getting a good mark.


As a starting point you might want to consider structuring your dissertation like this. However, bear in mind this is only general advice.
The advice on this page should be used in conjunction with the discussions we have as well as taking into account the instructions given by the members of staff who are responsible for the group and individual projects.

When writing your dissertation, please try to use headings and sub-headings, in order to split up the text. Also, use bullet and number lists where appropriate. these things make the dissertation much easier to read.

Interim Dissertation

In each of the chapters that you write, it is usually a good idea to have an introduction section which tells the reader what they can expect to read in that chapter. It is also worth having a summary section so which tells the reader the main points they should take away from the chapter. Using bullet points or a table might be a good way of summarising.

Abstract
This should summarise your dissertation. After reading the abstract the reader should know what problem you are tackling, the techniques you are using, the results you have achieved. The idea behind an abstract is that the reader should be able to read it in isolation and then decide if it is worth reading the rest of the dissertation. In the context of a third year project the reader should know what it is you are doing before they 'plough' through the rest of it.

Please note the abstract should summarise, not advertise. The person reading it should know what the problem is, the technique you are using and the results you achieved.

Introduction
This section should introduce the project. State what it is you are trying to achieve, why you chose this project and your motivation in doing it.
After reading this section, the reader should be in no doubts as to the project you are working on, why you chose this project and how you will judge if it has been a success or not.
If you do a good introduction as part of the interim dissertation you can simply copy and paste it into the final disertation.

Related Work
This section should describe any related work that you have found on the subject. You are really trying to put your work in context of what has been done before. This section is very important and will gain or lose a lot of marks for a good literature review. Some more advice with regard to your literature review (how to start it, what it should include, how to find suitable literature can be seen on the "starting advice" page.
If you do a good literature review as part of the interim dissertation you can simply copy and paste it into the final disertation.

Work Carried Out to Date
Describe the work you have carried out to date

System Design
If you have done any design, include it in the interim dissertation. It should include a discussion, plus any UML (Rational Rose, Booch, whatever) diagrams.

Testing and Results
This section will probably only appear in your final dissertation, but if you have any interim results you could report them here.

Work Plan for next Semester
Provide a plan the work that you intend to carryout between now and the end of the project. This should include timescales and milestones (e.g. the presentation date, the hand-in date for the final dissertation as well as any milestones you set, for example, development finished, testing complete etc.). You might want to include a plan as to how you have spent your time thus far.

Summary and Conclusions
You may not have much to say in the interim dissertation but you should still conclude as to how you think the project has progressed so far, what problems you have encountered and how you have overcome them. What problems you still have and what you plan to do to overcome them should also be reported.
You might also report how the project has changed since you first took it on and the reasons for that.

References
This is an important section. You should make sure you include a full list of references. It is closely related to your literature review.
Please do not simply have a list of web sites.

I, personally, will mark you down very heavily if your references are mainly web sites (I am talking ten's of percent).

Lots of web site references usually indicates very bad research, not least of all as there is no guarantee that the web site will exist for longer than five minutes. But, more importantly, the research community has not peer reviewed web sites. This is different to journal articles and conference proceedings and, to some extent, books.

The whole idea of a reference is to allow the reader, should they so wish, to access the material you are referencing. With this in mind you need to provide full details.

An example of each type of reference is given below, but you can fnd many more in any academic textbook or academic paper.

Book

Conference Journal

Final Dissertation

Note :

Once you have handed in your dissertation I would like you to supply me with an electronic copy. This allows me to put it on my web site so that other students can benefit from seeing it, in the same way that many of you benefitted by downloading previous students dissertation. You'll be aware that I password protect this area of my web site so that only people who I give the username/password to can access it. Ideally, I'd like the dissertation in a format that I can access easily (at the moment word is the preferred option).

I also require a text copy (EMAIL is fine) of your abstract. This allows me to update my web site much easier than if I have to type it in myself (especially if the dissertation is in a format that is not readily accessible, like PDF).


In each of the chapters that you write, it is usually a good idea to have an introduction section which tells the reader what they can expect to read in that chapter. It is also worth having a summary section so which tells the reader the main points they should take away from the chapter. Using bullet points or a table might be a good way of summarising.

Abstract
See above

Acknowledgents
This is an optional section and you would use it to acknowledge any help you received (e.g. from anybody you corresponded with who was particularly helpful, help you got from a company etc.)

Introduction and Related Work
See above

System Design
This section should describe the design of your system. It should include a discussion, plus any UML (Rational Rose, Booch, whatever) diagrams.

Implementation
This section should describe how you have implemented your system. For example, the testing you planned and carried out, any special data structures you have used, the problems you encountered and how you overcome them etc. etc.

Experiments and Results
This section should explain the experiments you undertook (with justification) along with the results. You should think very carefully about this section as you want to make sure that you tell a logical story and you present your results in the best way possible. Try and be methodical in planning and conducting your experiments and ensure that they are reported in a similar way. In addition, make sure that your analyse and critique your results - do not just present lots of data and expect the reader to understand and interpret it.

Suggested Developments
Suggest in this section how your project could be developed and taken forward.


Lessons Learnt and What I Would do Different Next Time
Do not be afraid to be critical of yourself (within reason). Part of carrying out a third year project is to learn from your mistakes by recognising them and show that you have learnt from them.

Time Management
In this section, you should out outline how you planned your time and how well you adhered to the plan. Be honest. Say what you underestimated, overestimated etc. A gannt chart might be useful, showing planned against actual.

Summary and Conclusions
Summarise your work and draw out the main conclusions


References
see interim dissertation

 

Appendices
Add whatever appendices you feel are necessary. It is worth putting in some of your code, but only a sample. Do not include all your code. It will not get read and it will just waste time and paper. We can see what your coding style is like from a couple of well chosen classes (assuming you are programming in an object oriented language such as C++.

If appropriate, it is also worth including a disc which contains the source code for your project as well as object code. This will allow me (or somebody else) to run the program and, if necessary, check that the system compiles.

 

Conference Submission

Writing a paper, based on your dissertation, that is suitable to send to a conference, is one way of increasing yor chances of getting a good mark (but I should emphasise that it is not the only way!). However, the benefit of having something that is ready to send (or better still having sent) to a conference says to your markers that you have produced something which is, potentially, publishable within the academic community. This is not a light undertaking and I will be highly critical of the paper that you propose to send to a conference and maybe even stop you submitting it. This is purely for academic reasons and may not have a bearing on your overall result as there are many factors that I have to consider before submitting to a conference.

I should emphasise that you should not send anything to a conference without my agreement as the work will have my name on it and I would not be happy to be associated with something which I do not consider to be suitable to send. Also, if the paper were to be accepted, then somebody (proobaly me) would have to go and present it. And I need to know what my commitments are, as well as being happy to present the work to an international academic audience.

If you are interested in writing a paper, which would be an appendix in your dissertation (and would be heavily publicised throughout) then speak to me. Do not just go ahead and do it yourself.

If you want to look at potentail conferences, two which would be good from the point of view that the call for paper dates are after you will complete your dissertation are CEC and ECAI. A list of other conference web sites can be seen on the ASAP Conference page.

 

Journal Submission

Everything I have said above (for conferences is just as valid for journals; just more some!! Journals publish the highest quality work and, as such, they are very tough to get into.

If you are interested in trying to publish your work in a journal, talk to me and we can discuss it in detail.

Potentail journals can be seen on the ASAP PhD page.


Links


EMAIL : gxk@cs.nott.ac.uk          Home page

Last Updated 27/08/99