 |
|
|
|
 |
|
||
 |
 |
 |
|
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
 |
|
||
Picking a research topic, be it for a Ph.D. thesis or class project, can be difficult. The best approach is to keep on reading papers and discussing them, and good ideas will pop up. Ocassionally, your advisor will have something that needs to be done and will assign you the problem to work on. But more frequently you will come up with something you think is a cool idea and would like to pursue it. There are several preparatory steps you need to take in both cases:
- Make sure this is an important problem and people care to see it solved, otherwise you will be working in vain. Try to distill this information from the relevant papers you read.
- Verify that your idea makes sense and looks feasible. Talk to other students in your group, to your advisor, other professors, etc. Get enough opinions that the idea is good, practical and can be done in the amount of time you have for this project.
- Make sure no one has done the same thing. Before starting any research, do the homework and make sure that no similar ideas have already been published. If they have, you need to find enough points of difference between them and your idea. You also need to make sure that your idea will work better than the existing approaches to solving the same problem. This is the ongoing process. As you work on your research, you need to keep an open eye for all similar/relevant approaches done by others, understand them and make sure your approach differs from them.
- Make sure you understand what you need to do in order to verify this idea. Do you have all the necessary information? Do you have all the equipment? If the research seems too demanding, consider involving other people in it.
- Ask yourself if you are excited enough about this work. If you are not, sometimes it is better to switch to another problem. For instance, if you are working on your Ph.D. and you discover after a few months or a year that you dislike the whole idea, it is better to change because you have several more years to go. If, on the other hand, you dislike the idea you are working on for a class project and a month has passed, you are probably better off sticking to it since the time is running short. Either way, figuring out early that you don't like some research topic will save you a lot of aggravation later. Your advisor might not be particularly happy if you switch interests all time, but he/she will appreciate if you share your doubts and will help you find a mutually agreable solution. Keep in mind that you can only do a good work on a problem that you are excited about. Otherwise, both you and your advisor will end up with a mediocre research that you don't want to boast with.
When doing research, think deeply about each important step in the project. Consider all options and choose the one that seems best suited for the problem. It helps also to make a note to yourself why you have chosen that option. It will be useful later when you write a paper and want to justify your solution design.
You will likely do several research projects during your graduate studies. It helps if you try to keep them related but diverse. Thus later, when you apply for jobs, people looking at your research project summary will gain an impression of a focused researcher with several strong interests. For instance, when choosing a class project, try to relate it to your current research rather than just choosing anything that comes to mind.
Short point summary
- Read a lot of papers and discuss them to find inspiration
- Make sure your ideas address an important, solvable but yet unsolved problem
- Be aware of how much you like/dislike working on the chosen problem and act on it
- Keep track of important project decisions
- Keep your projects tied together but diverse