Please provide an idea in your email of what YOU WANT TO DO. Please have some ideas of the details, rather than just generic area. e.g. if you want to create a computer card game, what game(s) are you thinking of? If you can't narrow it down then just suggest a few that work and we can discuss pros and cons.
I believe that motivation is the biggest problem that people have in the individual projects and that it needs to be something which you are interested in if you are going to finish.
The majority of the project mark is on the dissertation rather than the software. This means that you need a project that you can discuss if you want a good mark. Make sure that you can phrase it as "this was the problem I was trying to solve", "this is what I have done to solve it", "this was why it was a good way to do it", "I could have done it in these other ways instead but I chose this way because...".
You need to be able to put 'because' and justifications for what you did and why all of the way through your report if you want to do as well as you can. So don't choose a project where the solution is obvious and it's just a case of coding it up (i.e. time consuming rather than needing throught to do it).
I've supervised quite a lot of projects in the area of board games, card games or other simple games, and they seem to work well for students and allow them to produce something good which will get reasonable marks without too much problem.
I am interested in many types of board and card games. There are many things which could be done in this area and here are some ideas: You can find more information on a huge number of boardgames of all types here
I am particularly interested in people looking into AI players for traditional board games as well as some more modern games. There is a huge body of research into automated players for many traditional games, but there are any modern boardgames or card games which it could be interestingly applied to - giving an interesting reasonable-sized project with something new. This may, of course, possibly involve simplifying the rules a lot, or maybe not, depending on the student and the game of choice.
My research area is operational research so I am happy to supervise projects in that kind of area as well, but my experience has been that students tend to spend longer getting data, setting things up, etc, with that sort of project, so I am suggesting to go with the simple game ideas for time reasons, unless you are really a top student, you'll probably make something more impressive and get a better mark. These projects basically mean taking real world problems, building models of the problems and finding good solutions for them. As I mention, usually the problem with these is getting the right data to work from or getting the information to make an accurate model. In general the easy problems tend to have already been solved, and this project is relatively short for tackling a complex problem. But if you already have the problem in mind and know how to get the information and data then feel free to discuss with me.