WT!F?. How come posters above me know nothing about this? All these answers are incorrect. No wonder /sci/ is filled with bs. People here don't even have any MSc in STEM.
Here's how it should be done:
You talk to potential supervisors and see their research interests. They keep track of interesting projects that MSc students can do. Basically, they give you a problem and some ideas on how to solve it. Then, it is up to the student to investigate.
