dissertation ideas?

[stinkinrich88]

hello,

I'm doing a computer science degree at Cardiff uni and I have to decide on a dissertation topic in the next two weeks. That's pretty scary stuff. I'd love to hear if anyone has any good ideas to help me out.

Basically, I love programming in Java and C++. At the moment I'm thinking of using generics to optimise the travelling salesman problem. I'm aiming for a first so I hope this will have enough scope.

Does anyone have any cool ideas for my final year dissertation?

thanks!:KS

[renzokuken]

write a sick program to crack into Solus/The Taf and lower the prices even more !

[kripkenstein]

Well since you are here, I presume you like open source stuff :) Then one idea is to continue some research I saw on open source bugtracking/bug triaging. Basically this person used computer science methods and machine learning to decide how to triage bugs, at least in the sense of deciding to who to send them for initial inspection. This used the text in the bugreport and so forth.

So perhaps you could try that approach on other data, or variations on it, etc. I can find the original research article(s) if you want, but it should be easy to find using google.

Edit: Added link per request, in a post below.

[23meg]

Basically this person used computer science methods and machine learning to decide how to triage bugs, at least in the sense of deciding to who to send them for initial inspection. This used the text in the bugreport and so forth.

I'd appreciate the link; searches didn't yield a result.

[kripkenstein]

Do you have a link?
Sure, here:

www.cs.ubc.ca/labs/spl/projects/bugTriage/papers/icse2006.pdf