Premise: You are in no need for batch software installation.
I would say why not build a tool which you can use in your daily life. And if you find it useful, tell other hackers about it and see if it makes anyone else's life easier too.
This way you can learn the non-technical aspect of product development too (design, usability etc.)
The batch software installation idea was not really my first choice. I was thinking more about doing a web application - had many ideas already in place - but this wretched 'no database' restriction has ruined nearly every classmates' plans including myself.
The point is, I find it quite difficult to think of a project not involving a database. That's one of the main reasons I'm asking for advice here.
I really had hoped to do something spectacular for my final year project.
To reiterate what others are saying, you can do web stuff without necessarily relying on a database.
Implementing computer vision techniques in Javascript seems to be quite popular at the moment and could make for a good project. You could get a lot of mileage out of exploring what techniques are feasible and which aren't and what you can do to optimise things. Depending on what your eventual application is it will likely need a user interface - make sure you back up its design with HCI theory and methodical evaluation and that will also add to your grade.
I would say why not build a tool which you can use in your daily life. And if you find it useful, tell other hackers about it and see if it makes anyone else's life easier too.
This way you can learn the non-technical aspect of product development too (design, usability etc.)