PDA

View Full Version : Final year project



khpuce
06-10-2003, 05:04 AM
Hi everybody. I am gonna start my final year in BSc Software Design & Networks in September. During my final year, I am supposed to do an individual project releted to my subject. But I am really struggling to find a good, suitable project to do. Though my course covers both softwares and networks, I am interested to do something on softwares. But the problem is, the only thing I can think of software is creating some sort of database program which I feel is not good enough for a final year project. Can anyone suggest a topic for my project?

Thank you all.

SourceCode
06-10-2003, 08:12 AM
You could make your project beyond good, make it one of a kind make it the best. Make an irc client that as a bonus that allows it's users to connect to Fast Track, it has over 3.5million users on it. That's right the network kazaa runs on. gIFT does this now I think they broke the encryption very very recently and it's open source so you could simply look at the plugin and see how they broke it, and of course give them credit etc. The problem is gIFT is so freaking unstable, and the openft network it uses is always in development and just buggy. It's not very user friendly at all, noobies simply can't use it. I gaurantee that 80% of the people who download gIFT don't get through the install and if they do they configure something incorrectly etc. Your program could change all that. You would provide a way for linux users, especially people who aren't to linux savy to finally download stuff that kazaa users can and at the same time allow them to hang on irc and download more stuff. The ultimate filesharing program. It would definitely not only get you an A but it would be a very popular program. You would need to read of up on the irc protocol and of course, how to break the encryption that Fast Track uses.. Goodluck.

khpuce
06-10-2003, 08:27 AM
Thanks for your advice SourceCode. Though most of what you wrote seems alien language to me at the moment, I'll definitely look into this and try to get more information about irc. But I am not sure whether I can manage it within a very specific time limit because I am not amazingly good in programming and consider myself as an intermediate level programmer.

Thank you very much again and I would very much appreciate any further information about the whole thing.

iain
06-13-2003, 03:44 AM
What im thinking of as a finaly year project is a peer - to - peer filesharing system with cataloged content. You can then include:

File i/o
networking
protocol negotiation
robust communication
security
databases
database access issues (locking, concurrency etc)

there are endless possibilities with this project

khpuce
06-13-2003, 05:28 AM
Thanks iain for your nice suggestion. I think this is very much achievable with the time limit. Would you please explain the unique benifit of this project.

Thanks.

N.B. To everyone: At the moment, I am gathering as much ideas as I can (without much success:( ). So, whoever watches this post, if you have an idea please let me know (All I am just asking for ideas, not any specific solutions!!!).

vasanth
06-13-2003, 07:29 AM
well how about a game with network support..

iain
06-13-2003, 08:44 AM
>>Would you please explain the unique benifit of this project.

The benefits are that it brings together all the most important components of computing and combines virtually all the material, the good thing is you can specialise in your favourite area.

So if networking is your area, concentrate on a client/server system, develop your own protocol

If security - develop a firewall, basic file integrity tests etc

The best thing is, it can be written in virtually any language, with any database query language.

It provides enough material for research and enough practical work so you dont get bored with reasearch. It is also perfect for the dissitation time period.

khpuce
06-15-2003, 04:53 AM
Thank you very much for the clarification. Now I am getting the picture. Thanks again and have a nice time. :)

khpuce
06-15-2003, 04:58 AM
A game with network support....Ummm....That sounds really good (and easy perhaps!).

crag2804
06-15-2003, 07:06 AM
I think the idea of a genetic algorithm problem solver sounds like a really good idea. Using "natural selection" you could have it solve some problem like exam timetable scheduling.

khpuce
06-15-2003, 07:14 AM
What is a genetic algorithm problem solver? It REALLY sounds C..O..O..L....But does it involves lots of maths?

Zach L.
06-15-2003, 09:39 AM
A GA finds (nearly) optimal solutions to optimization problems by generating a set of solutions, evaluating their fitness, and then breeding solutions to form a new population of solutions (it evolves solutions). Mutation of some sort is often involved as well.

Check here (http://www.generation5.org/) for some more info.

iain
06-17-2003, 06:18 AM
iBut does it involves lots of maths?

it contains more maths than you could imagine, and the horrible kind too!

khaled20
06-18-2003, 05:00 AM
Hi everyone,

I talked with one of my tutors and considering the situation (course relavance, requirements, feasibility etc.) we selected two of the ideas.

These are :

1. A game with network support (Thanks to vasanth)

2. A genetic algorithm problem solver (Thanks to iain)

From the above two, I personally prefer the GA. I will try to find out exactly how much maths are involved in the GA and if I think I can handle the load then I will do it otherwise my 2nd choice (at the moment!) is the game with network support.

These are the ideas I am considering at the moment. I am still looking for ideas and may change these, if I come up with more interesting ideas.

Thanks to EVERYONE for your opinions/suggestions. You guys are geniuses.

khpuce
06-18-2003, 05:21 AM
Whooops....I think I've just made a mistake. The last post is showing its author as khaled20 (guest). Actually it was me.

Sorry for the inconvanience.

Brian
06-18-2003, 06:31 AM
Write your own open-source minix kernel. In QBASIC.

khpuce
06-19-2003, 03:19 AM
Thank you very much Brian for your suggestion. First I'm not sure what an "open-source minix kernel" is and second, I am not really a QBasic fan. I intend to do my project with C/C++ or java.

Thanks again for your reply.

N.B. Does anyone here knows anything about s/w OPnet modeller ? One of my tutors has suggested me this. Then including:

Effects of QoS on application response
Use of ACE

To be honest, I don't have a clue what he was talking about. :confused: Perhaps any help from you guys?

khpuce
07-18-2003, 06:08 AM
Hello again. At the moment, I am thinking of creating some sort of restaurant / Clinic management software. But I am not sure which programming language to use for this ( i.e. C, C++ or Java ? )

Any help from you guys please ???

Thanks.

Zach L.
07-18-2003, 10:11 AM
I'm not entirely sure what you mean by management software, but it sounds like something that a) needs a nice user interface, and b) doesn't need to be extraordinarily fast (no extremely compute intensive calculations). With that in mind, I'd say Java.

khpuce
07-18-2003, 09:09 PM
You've got the idea PERFECTLY Zach L. Thank you very much for your suggestion.

khpuce
10-06-2003, 06:18 AM
URGENT PLEASE!!!

Could anyone tell me what kind of maths are involved a GA problem solver? (e.g. Calculus, Trigonometry etc...)

I will be grateful if anyone gives the answer ASAP.

Thank you.

Raihana
10-07-2003, 07:18 AM
I am not sure but I think if you want to create something like an exam timetable scheduling, I don't think you have to use genetic algorithms instead you can use simple logic and arithmetics. Can anyone tell me whether I am right or wrong?

khpuce
10-10-2003, 07:04 AM
I just want to say that, finally I am doing my project on creating an automatic timetable scheduling program (Many thanks to crag 2804 :)

I will try to do it using Genetic Algorithms (if I can understand and manage GA !!! ) otherwise I'll try to develop my own algorithm.

Thanks again to everyone for your kind help.