Applicants should normally have completed an M.Sc. in Computing Science or a closely related field at an established university. They are admitted in accordance with the general regulations of the School of Graduate Studies. Candidates normally enter the program at the beginning of the Fall term (September).
Candidates pass through six stages of activity:
Candidates are encouraged to obtain a supervisor from within the School prior to commencing their studies. Any member of the School of Graduate Studies may supervise a student. A candidate may be jointly supervised by more than one person. In cases of co-supervision, one supervisor may be from outside the School.
The Ph.D. program is managed by the Ph.D. program committee who are responsible for: maintaining consistency in the requirements for individual candidates, approving the candidate's progress through various milestones, handling disputes that may arise, and acting as chairs and examiners when needed. The Ph.D. program committee is appointed by the School Director. The Ph.D. program committee reports annually to the Director on the progress of each Ph.D. student.
All candidates are under the direction of a supervising committee consisting of two faculty members and the supervisor or the co-supervisors. At most one of the committee members may be from outside the School. The candidate obtains a supervisor during the first month of the PhD program. A research topic is agreed to by the candidate and the supervisor.
The candidate and supervisor agree on possible members of the supervisory committee whom they invite to serve. When a willing committee has been formed, its make-up is sent to the Ph.D. program committee for approval. The supervisory committee should be selected and approved within the first two months of the program. The supervisory committee can be changed at any time at the request of the student with the concurrence of the Ph.D. program committee.
The Ph.D. program committee appoints one of its members to act as the liaison between the supervisory committee and the School. This person is available to advise the candidate and supervisory committee on School procedures and to mediate disputes within the supervisory committee.
The supervisor directs the candidate's day-to-day work and acts as primary resource for the candidate. The other members act as a secondary resource. The supervisory committee must approve the breadth requirements and research topic proposal documents before they are submitted to the PhD committee. The supervisory committee normally meets frequently with the candidate. Three times a year it reports on the candidate's progress to the Ph.D. program committee.
There are four formal requirements imposed by the School: a breadth requirement (which can be met through a combination of graduate courses, an MSc thesis, and work experience); a six-page topic proposal; a comprehensive examination (which includes a written research proposal and an oral defense); and a Ph.D. thesis. Candidates who fail to complete any requirement within the time limit specified will be required to withdraw from the program. The Ph.D. program committee may approve extension to time limits, subject to the rules of the School of Graduate Studies, and will normally do so for part-time and inactive students.
A candidate for the Ph.D. degree should demonstrate broad knowledge of the various areas of Computing Science, and in particular, the way in which these areas relate to one other.A token-based approach is used to evaluate the candidate's knowledge. A token represents knowledge of an area such as might reasonably form a graduate course. The following activities would normally count as tokens: a grade of at least B- in a graduate course at Queen's (excluding CISC 897 & CISC 810) or equivalent in standard to one at Queen's, a research project, appropriate work carried out in industry, or a Master's thesis.
The candidate demonstrates breadth by presenting ten tokens with an appropriate distribution among the following areas:
Area 4 (Multidisciplinary Studies) is optional and is intended to encourage recruitment of students who may not have a "conventional" computing background but who clearly demonstrate the skills and methodology that we expect of our doctoral students. At least two tokens in each of the core computing areas (theory of computation, computer systems and applications within computing) are required. The School of Computing Graduate Handbook classifies graduate computing courses under the three core areas and therefore gives guidance about how a token in each area can be obtained. Normally, at least 4 courses from the School of Computing are required. The Ph.D. program committee ultimately decides whether a proposed token is acceptable.
When a supervisory committee has been appointed, the candidate presents a plan for meeting the breadth requirements to it for discussion and approval. The proposal must provide a clear description of each token. Courses require calendar description and name of instructor, a thesis token requires title and abstract of thesis and a work token requires job description.
When breadth proposal has been agreed upon, it is sent to the Ph.D. program committee for approval or modification. The Ph.D. program committee ensures that similar requirements are applied to all students.
Where a token is failed, a candidate may appeal to the Ph.D. program committee to substitute another token or to repeat the failed requirements.
The breadth proposal must be approved by the supervisory committee and submitted to the PhD committee before the end of the first term into the program. The breadth requirement must be fulfilled within one year of the first registration to maintain satisfactory progress.
The second School requirement is a 6-page maximum (Queen’s thesis format not including references) topic proposal. This document is to be submitted before the end of the third term after the initial registration in the Ph.D. program. Failure to submit the proposal on time is considered unsatisfactory progress.
The document gives a description of a subject area (one paragraph, roughly comparable to a course description), and a proposed topic for research to be conducted within that area. The document then provides a brief survey of the key and most pertinent literature contributions for the proposed topic.
The Ph.D. supervisory committee provides feedback on the document and their expectation for the Ph.D. Research proposal. The feedback from the supervisory committee gives the expected weights of the literature survey and research plans in the PhD research proposal paper. The feedback is communicated to the Ph.D. committee which in turn communicates it to the student.
Graduate students enrolled in the Ph.D. program in Computing at Queen's are required to pass a Comprehensive Examination as described in Calendar of the School of Graduate Studies.
The examination ensures that candidates are well-versed of the state of the art in the area in which they intend to pursue research, that they have a critical perspective of the area, and that they are able to formulate a research plan to explore open problems and research opportunities. Candidates should have acquired their own views of the area, be able to be critical of previous work, be able to discuss the area with other researchers at their own level, and able to outline a well-thought research plan. Although they may not themselves have contributed to the area, they should be as informed and analytical as those who work in it.
The proposal requirement is met by writing a PhD research proposal paper (as defined below), presenting it orally to an examining committee, and answering questions about the proposed research and the associated area.
The PhD research proposal paper is a 40-page maximum (Queen’s thesis format not including references) research proposal. This document is to be submitted before the end of the sixth term after the initial registration in the Ph.D. program. The document should normally cover background material (e.g., motivation and literature survey), the problem to be tackled, methods to be used (e.g., research plans and experimental design), results sought, evaluation metrics (i.e. how research success will be measured), and milestones (including progress to date). The proportions of text for the literature survey and for the research plans will have already been communicated to the student through the feedback received on their topic proposal.
When the candidate is ready for the examination, he or she asks the Ph.D. program committee, through the supervisor, to schedule the exam. The supervisor is responsible for finding a suitable examiner. The graduate coordinator appoints a Chair (normally a member of the PhD program committee) and schedules a time for the examination. A copy of the research proposal paper must be delivered to the members of the examination committee by the candidate (Chair, members of the supervisory committee, and examiner) at least two weeks prior to the scheduled exam. Members of the examination committee who are absent may participate by submitting questions in writing. These questions are put to the candidate by the Chair.
At the exam, after a closed meeting of the examining committee, the candidate gives a 20-minute presentation of material. Members of the examining committee will then address questions to the candidate that may cover all aspects of the research area and the proposed research. The examiners may ask questions that allow them to judge the candidate's comprehension of the research area, to assess the candidate's ability to undertake the proposed research, and to evaluate the ability of the candidate to defend his or her claims from the research proposal.
The Chair is responsible for the conduct of the meeting and does not ask questions (other than those of absent committee members).
2. OUTCOME OF THE EXAMINATION
After the questioning, the examining committee meets and reaches one of the following three conclusions:
a) The candidate has passed the comprehensive examination. b) The candidate has passed the comprehensive examination but the committee has significant concerns about either the candidate's mastery of the research area or the quality of the proposed research plans The Chair writes the candidate a letter outlining these concerns. The letter may include a list of corrections/modifications for the research proposal paper that are a required for passing the examination. In this case the letter also specifies members of the examining committee who will verify that the changes have been satisfactorily implemented. c) The candidate has not convinced the examining committee of his or her deep understanding of the area and/or his research plans, and has failed the comprehensive examination. A second (and final) attempt is allowed.
The Chair writes to the candidate to this effect, and invites him or her to suggest a research proposal for a second comprehensive examination (not necessarily a different proposal). Once the research proposal has been agreed upon, the procedure above is repeated. The examining committee normally puts a strict time limit on the second (i.e., final) attempt of the comprehensive examination.
The committee's decision on the proposal examination is based on the quality of the written document, the quality of the candidate's oral presentation, and on the candidate's response to questioning during the comprehensive examination. The committee can ask questions belonging to the general thesis subject area and topic, as well as, questions on the proposed research and outlined research plans. The Chair is not a voting member of the examining committee and the Chair can use her or his discretion on whether or not to read the proposal paper before the examination.
If the committee does not reach a unanimous opinion, the committee votes on the outcome. A decision to pass the candidate on the comprehensive examination can have at most one dissenting vote. The Chair reports the result of the comprehensive examination to the Ph.D. program committee. If the comprehensive examination is passed, the final document, integrating all examination feedback, is kept in the School library. The comprehensive examination must be passed before the end of the sixth term after the initial registration in the Ph.D. program.
3. DELAYS IN EXAMINATION
The PhD research proposal document is normally expected to be submitted before the end of the fifth term after initial registration in the PhD program. If the PhD research proposal document is not submitted before the end of the sixth term, the student will receive an automatic failure -- leaving him or her with only one final attempt to pass the Ph.D. Research Proposal requirement. The Ph.D. supervisory committee needs to schedule the final attempt (normally within 1-2 terms), once the automatic failure has occurred.
4. PROGRESS REPORT
Within one year of defending the Ph.D. Research proposal, the student should submit a brief 6-page maximum progress report (Queen’s thesis format not including references). The report should summarize the progress over the past year and any significant changes to the plan that was detailed in the proposal (along with updates to the milestones). Failure to submit the progress report on time will lead to the student receiving unsatisfactory progress.
5. APPEAL OF A FAILED EXAMINATION
5a) If a student wishes to appeal the outcome of a comprehensive examination on procedural and/or academic grounds, an appeal must be lodged formally with the Coordinator of Graduate Studies in the School of Computing. The appeal should explain in writing why the student believes the academic decision is unjust. This should be done as early as possible, and normally not more than ten working days after the comprehensive examination. The Graduate Coordinator must respond to the appeal within two weeks of receiving the appeal.
5b) If the matter has not been resolved by the Graduate Coordinator and the student continues to believe that the academic decision is unjust, a formal request may be lodged for a review of the appeal by the faculty members of the Graduate Committee.
The faculty members evaluating the review shall not include members of the examination committee of the student's comprehensive examination.
When conducting the review of the appeal, the Graduate Committee shall interview the student and the members of the examining committee from the comprehensive examination. The Graduate Committee may find either that
(i) The decision of the comprehensive examination is academically and procedurally sound and the appeal is denied.
(ii) An error in procedure or academic judgment has been made. In this case the Graduate Committee shall proceed to rectify the error. Normally this would mean that the student is given a new attempt at the comprehensive examination. If the appeal deals with a first attempt, the new attempt is considered to be the first.
The Graduate Committee also has the option of changing the outcome of the comprehensive examination from failed to passed.
5c) If the Graduate Committee finds that a negative decision for a second attempt at the comprehensive examination is academically and procedurally sound, the negative decision has to be approved by the Director. If after a second failed attempt the Director and the Graduate Committee recommend to the Chair of the Science Graduate Council that the student will be required to withdraw on academic grounds: the student may appeal the recommendation to withdraw by following the procedures outlined in the School of Graduate studies calendar.
A thesis presenting original research is submitted for approval by a thesis examining committee. The research described in the thesis should constitute a significant contribution to knowledge in an area of Computing Science. It must be original and of such quality as to warrant its publication in a recognized journal.
Candidates should consult the Calendar of the School of Graduate Studies and Research for University Regulations regarding Ph.D. theses. The thesis is examined in accordance with the general rules of the School of Graduate Studies and Research.
Any of the following are grounds for requiring a candidate to withdraw. The list is illustrative rather than exhaustive.