1 Introduction

This document contains some personal advice on finding a graduate supervisor in the Faculty of Computer Science at UNB. It does not represent Faculty policy, only suggestions by the author. Overall, finding a supervisor, whether or not there is funding involved, is a great deal like finding a job. You need to find a project that you can be enthusiastic about, and you need to convince a supervisor that having you as a student will be rewarding, i.e. that you have something to contribute. The rest of this document is divided into two sections. The first discusses how to choose some possible supervisors to approach. The second discusses how to convince someone to take you on as a student.

2 Choosing candidate supervisors

If you are reading this document, then the chances are either that what you put down as your research interests in your application are not closely related to those of a professor, or they are, but those professors are swamped with students. So you have to work a little harder. I recommend making a list of five professors who you consider possible supervisors. The first step is find out about professors.

2.1 Take a course

The best way to find out what a professor is like, and what their research is like, is to take a course from them. I know from my own experience as a professor that I have often been impressed with students in my classes whose applications did not make much of an impression on me. In a course you will discover not only what the topics really mean, but what how the professor thinks. If you are a practical nuts and bolts person, you may not want to work with someone who only likes to prove theorems. Contrawise, if you really like proving theorems, you will probably not be happy working for someone who only likes to program. Don’t forget that it is sometimes possible to take a “reading course” on research topics.

2.2 Read web pages and office doors

Many professors have information about their research and potential student projects on their web pages. Others prefer to keep discussion of the projects within by posting them on office doors. Some of these projects are very precisely defined, others are more flexible. Don’t be afraid to ask professors questions about their projects. Mostly the easiest way to get in touch with professors is by email.

2.3 Read papers, or at least abstracts

Many professors have publications either on their web site or on their office door. You can get a good idea about a professors research by reading some of these papers. If the papers are long, then you should at least try reading the abstracts. If you cannot find publications on the web, or posted on doors, try the electronic indices in the library. The librarians in the Engineering library will be happy to help you use the electronic indices.

2.4 Don’t forget NRC and Saint John Campus

Several researchers at NRC are adjunct professors at UNB. This means they can supervise graduate students, but only if you have a co-supervisor from UNB. Usually it is much easier to find such a co-supervisor than to find a regular supervisor. To have a supervisor from Saint John, no other supervisor is necessary.

2.5 Think interdisciplinary

When you are looking at projects, think about how you might combine your special interests with those of the professor. Many professors are open to proposals of research projects, but remember most of have received research funding for a specific program of research, thus whatever graduate students we fund must advance this program.

3 Making the sale

Supervising a graduate student takes a great deal of work (although not as much work as being a graduate student). In some cases you will also be receiving research assistantships (funding) from a professor. Thus you need to be able to answer the question “What is in it for the professor?”, i.e. what can you in particular contribute to the research project.

3.1 Know your professor

See Section ??. There is two main reasons for this. The first is that if you know nothing about a professor’s research, then the probability of a good match in interests is small. The second is that you will appear much more clever if you can talk in some depth about a project. On the other hand, beware of outright flattery. Many professors do not react as well as you might expect to to “I would like to work with you because you are the top person in the field”. Generally we know who the top people in various fields are, and if it is not us, then that just reveals you have not done your homework.

3.2 Know yourself

For each professor on your list, think about what particular skills and experience you have that is relevant. Being enthusiastic about an area is a good start, but having previous coursework or industrial experience is much better. It may be as general as “I am good ’C’ programmer”, or as specific as “I am really interested hardware implementations of graph algorithms”.

3.3 Get backup

Academics are naturally skeptical. This is nothing personal, but a necessary side-effect of being a scientist or an engineer. You may have all kinds of great reasons why Professor X should take you on as a student, and maybe you can convince them just by talking. But why take a chance. You should also consider providing

1. Your UNB transcripts.
2. A short resume (two pages maximum) explaining your relevant experience.
3. References from other professors. If you, for example took a course from a professor, did well, but for whatever reason do not end up working with them on a thesis or project, then consider asking them to be a reference for you. Graduate courses are the most convincing here, although personally your professor from 3913 would also be convincing.

4 Good Luck

If you have any questions or comments on this document, please let me know. I will try to update it as circumstances change.