Erm.. been hit with a bombshell.

The original idea of grid computing over mobile devices is no longer a valid final year project idea for me because of this lovely paragraph I got in an email.

The project you choose will also depend on your choice of programme. If you choose a Computer Science with programme your project must match your with option.

My "with" option is Knowledge and information systems.

Which to me screams "DATABASES!!!!" but thats about all I know.

Currently stuck with ideas revolving around a CRM suite or a website of somekind :| but I have this week to come up with a project proposal, and I am stuck. Really dont want to do a CRM suite, did one for A level... its boring. Any suggestions from you fine lads and lasses there that I can take for a project idea?

Needs to be associated with "knowledge and information systems" however that is interpreted by yourselves.

Serious, the vibrations on the phone are not to be used in that manner!!!

it screams data to me also but I just dont know how knowledge systems come into it.

BTW BigSoy is there any kinda explanation you can give me with regards to knowledge and information systems? As I dont truely know what it is. I picked it as a large number of the modules are to do with databases and associated subjects. But Ive no real idea of the meaning of it.

Got to say that back in the day (god I sound old) Knowledge Based Systems was a secondary topic within AI.

Where KBS focused on systems that made decisions based on previously known data.

AI being those systems that make their own decisions.

Obviously the two overlap when the AI system can store its decisions and results, pulling on those results to later determine the best possible solution to a new problem, or indeed ignoring the results and attempting a new solution.

Yeap, thats about the size of it.
The KB system can grow its *database* such that its answers become more relevant. e.g A simple search engine provides better results with the more webpages it indexs.

An AI based system will improve its reponsiveness through deduction / feedback of the given results to a problem. e.g Finding your way through a maze, each time a decision leads to a dead end the system will go back and try again. Eventually the system will learn that following one wall will always lead to the exit.

Then there are the highbred systems that use both a KB and AI to improve the system. e.g. Google, it uses both the KB of all of the indexed pages along with the feedback of "click throughs" to determine which pages are most relevant to a given search string.


I will have a little think about projects but neither of these subjects are ones I have much experience with (I steered well clear of AI and KBS as I hated the maths )

Aye already studied that in maths Along with string matching and so on.

I was thinking of a spamfilter but for what?! theres no reason to develop another spam filter for email as the market is flooded anyway.

I am also thinking of a search engine but I dont know how to base it. Its very very easy to make a search engine script input pages into a database (too easy for a final year project, its a 1st year assignment), but as nimrod says it goes into the realm of AI if the system learns from clickthroughs or votes or feedback of any kind