A description of how you propose to undertake the evaluation of your work. You must ultimately be able to answer the question of whether or not the work you have proposed and (later) completed is important. This is often done by direct comparison with other, existing work in the field. Such comparisons may be done experimentally, analytically, through simulation or possibly a combination of these. For example, you might be proposing a thesis where, at the end, you will want to compare the performance of an algorithm you developed and implemented with the performance of a similar existing algorithm. When doing this, always try to make the comparison(s) as objective and meaningful as possible. Compare your type of apple to someone else’s type of apple, not to an orange. Be sure to explain the methodology behind your comparison (e.g. how you will gather performance results accurately or how you will construct and run a simulation study). Always remember to keep statistical significance in mind whenever this is appropriate. Results based on samples of small size do not constitute evidence of improvement nor do results where the degree of improvement exceeds the margin for error in the experiment. When actually doing the assessment, try to be totally objective and always resist the temptation to tweak your work until you get the “expected” results. Instead, explain the results you get.
A statement of the resources required, if any, to complete your work and a description of where you will gain access to these resources. For example, if you need to have dedicated access to a number of machines for an extended period of time, then you need to say that you have the agreement of the machines’ owner(s).
A tabular set of timelines that provide realistic estimates of when the major phases of your thesis will be completed (including the writing of your thesis). These are often difficult to predict without experience so be sure to involve your advisor in setting these dates. Remember that it almost always takes longer than you expect to get anything done and that you will likely also have other responsibilities (e.g. coursework, marking, etc.) while you are trying to complete your thesis.
A brief (one to two paragraph) summary of the proposal (i.e. the previous sections) that highlights the key points in the proposal and provides a list of contributions to the field that you expect your work to provide. Be very specific when listing your contributions and explain why they are of interest to the computing community.
A bibliography of the papers, etc. you have read and cited in your proposal. The bibliography should be ordered in a convenient way, normally by last name of first author and should use a consistent style for all entries. (Note that using LaTeX and BibTeX is an easy way to ensure such consistency and ordering.) Each entry should contain complete information (e.g. not be missing page numbers, etc.). The selected papers in your bibliography should be carefully chosen to be up to date, important references in the field.
6. The review process and the GSC’s response
A minimum of three members of the GSC, selected by the chair, will provide reviews of your proposal to the chair. The committee as a whole will consider the comments of the reviewing members and will place your thesis proposal in one of three categories:
Last modified: 27/02/2007
Page 5 of 8