[hazelrink]

Okay unfortunately i got my proposals rejected at school...

here is what i requested and why they werent approved...
1. a program that has a music manager connected in the internet has a notepad check your horoscope basically like a OS or its like you opened your itouch and your playing with it...

reason: my prof said it was like an OS and people dont need it...

2. Learning Management System - a web service/software for student and teachers... you could upload educational videos, see your schedule, grades, news on the school, take exams online and many more... for the professor if she/he uploads the grade of her student the client automatically texts and emails it to the student

reason: she said this is the kind of software they're looking for, for IS and IT students not CS students... T_T

3. chat server

reason: there are too many chat servers already and she thinks i cant do it hahahaha

4. English to any other language converter - you input a paragraph in english and the program would translate it

reason: too many groups that has already proposed it



this is what she saw feasible for us but has a problem


1. web robots, web crawler, search engine

what she said: we need to improve the algorithm of the search mechanism and she asked me why do we need to improve web crawlers is google not enough?

2. bar code generator

what she said: it is possible but why would you make a bar code generator is there something wrong with the existing bar codes?








SO basically i need to find an algorithm which needs improvements... Thanks for those who are willing to give me ideas T_T

[William_Wilson]

What exactly do you mean by an algorithm that needs improvement? Are you being asked to create software that is better than everything that already exists? If so, that is an insane assignment!

Google's search algorithms are custom tailored since they are run on Google OS, with a Google File System over countless machines using a form of map reduce methodology to return results as quickly as they do.

People are always trying to improve sorting algorithms, especially those that run on GPU hardware and distributed systems.


Can you be more specific about the scope of project you are looking for?

[hazelrink]

William_Wilson, on 25 Nov, 2009 - 05:41 AM, said:

What exactly do you mean by an algorithm that needs improvement? Are you being asked to create software that is better than everything that already exists? If so, that is an insane assignment!

Google's search algorithms are custom tailored since they are run on Google OS, with a Google File System over countless machines using a form of map reduce methodology to return results as quickly as they do.

People are always trying to improve sorting algorithms, especially those that run on GPU hardware and distributed systems.


Can you be more specific about the scope of project you are looking for?

actually im really a fool for picking web crawler even though i have no idea how a google, yahoo or any search engines work... i just now they use metadata (or do they?) i was just inspired to do a web crawler because i found a good book on how to start a search engine program...

we need to a program that has some use to the society... a program that has a sense...

here are samples what proposals they approved...

A.I. for chess or game of the generals - the problem is they need to prove that their algo is better than the existing one o_O
thats overkill

Image enlarger software

team viewer - something like this

they also approved decrypting, encrypting

compression and extraction

shortest path... Map integrated ... GPS o_O

English to any language converter

Math problem solver (trigo, calculus no algebra... and it needs to be specific)



I was also thinking about improving a sorting algorithm of a software but im not too confident with it

[Momerath]

The problem I see with all the projects you listed is that people have spent years of their lives trying to improve these things, and fail. For this to be a class project is silly.

[SpeedisaVirus]

yeah, I wouldn't want to undertake this as a project. Anything that has a reasonably easy means to improve has already been improved. Like Momerath has said, people far more brilliant than most of us have worked on problems like this for years to not find an improved algorithm.

Is this like a senior capstone project or is this something lower?

Here are a few of my ideas:

Multi client IM program. I haven't done one of these but it seems easy enough.

File sharing program. Writing a torrent app isn't too tough but then you could add an FTP feature to it or something to make it a more complete data transfer package. Adding things like bandwidth graphs and etc would be extra bonus cool.

An actually effective bulk ID3 tagger and music organizer. It could reference online databases to propose possible tag data. Then if that is accepted you would then rename the files and organize as needed.

Simple to use image editor? Think Paint.NET or something similar. That's really easy on the surface but implementing some of the more advanced features could be fun.

I know one group is doing encryption but if they are doing symmetric key encryption maybe you could do asymmetric key encryption. I don't know how challenging that would be but there is a lot of information out there on how it works.

This post has been edited by SpeedisaVirus: 25 November 2009 - 01:36 PM

[hazelrink]

Momerath, on 25 Nov, 2009 - 11:03 AM, said:

The problem I see with all the projects you listed is that people have spent years of their lives trying to improve these things, and fail. For this to be a class project is silly.

SpeedisaVirus, on 25 Nov, 2009 - 12:33 PM, said:

yeah, I wouldn't want to undertake this as a project. Anything that has a reasonably easy means to improve has already been improved. Like Momerath has said, people far more brilliant than most of us have worked on problems like this for years to not find an improved algorithm.

Is this like a senior capstone project or is this something lower?

Here are a few of my ideas:

Multi client IM program. I haven't done one of these but it seems easy enough.

File sharing program. Writing a torrent app isn't too tough but then you could add an FTP feature to it or something to make it a more complete data transfer package. Adding things like bandwidth graphs and etc would be extra bonus cool.

An actually effective bulk ID3 tagger and music organizer. It could reference online databases to propose possible tag data. Then if that is accepted you would then rename the files and organize as needed.

Simple to use image editor? Think Paint.NET or something similar. That's really easy on the surface but implementing some of the more advanced features could be fun.

I know one group is doing encryption but if they are doing symmetric key encryption maybe you could do asymmetric key encryption. I don't know how challenging that would be but there is a lot of information out there on how it works.

actually this is a thesis... sorry i didnt mention it...

i got my ass kicked again... just a few hours ago T__T
heres what i proposed:

1. Key Logger
reason for rejection: she said it was unethical... I said no ma'am it depends on how the person use it she said: how? I said: monitoring the activities of the user that is using the computer she said: the os could do that... END of conversation but really i dont know how the OS could monitor the user's activity...

2. Restriction Program: like restricting and putting a password to a software so that not anybody can open and use it... she said: the OS could also do that and it seems like she really doesnt want the idea...

and i suggested another but cant remember it im too tired...


actually i told her that our thesis is hard... they want us to find an algorithm to improve for example compressing a file what they want us to do is to get the algorithm for compressing and improve it but how the hell would we do that is the big companies havent improved it or cant improved it how the hell will we able to do it?

but then she said CS thesis is really hard... your scientists your thesis level is higher than the IS and ITs.... i really regret choosing CS since i love database programming but i cant do it since thats for the IT and IS students T_T


oh and she suggested for me to find topics that are being talked about right now read peoples comment on the topic and recommendations on what fields the software programming is lacking and needs improvement but i cant find any...

oh and i also need journals and books references for the topic i would be choosing before they approve it... (she said it would be better if i got the algorithm already)

im so depressed...

if i fail my thesis 1 this sem then this would be the first subject i would fail for the past 2 1/2 years

oh and ill try researching about that symmetric and asymmetric encryption

This post has been edited by hazelrink: 25 November 2009 - 11:18 PM

[hazelrink]

oh my classmate just told me a technique to fool my professor hahaha he said i just need to search an algorithm and think of an application where i could implement that algorithm...

I have found one Rabin-Karp Algorithm for hashing and mainly for searching a word! ahahaha i could use the algo to make a program for checking whether some article was plagiarized or not or searching for a word

any other ideas guys?

[mostyfriedman]

if that's what you would like to do then go ahead and propose it, however there are programs that already do that , so you may get the same reply from your prof

[hazelrink]

mostyfriedman, on 26 Nov, 2009 - 03:12 AM, said:

if that's what you would like to do then go ahead and propose it, however there are programs that already do that , so you may get the same reply from your prof

yeah your right... and know what? i just found a thesis exactly like what i have said... but the good thing is, maybe i could try improving their work since its not 100% accurate

though im still searching for more algorithm that i could use...

and its very tempting to go cryptography since i find cryptography very lively in the net there are lots of journals and related literature about it! but... the problem is there are too many students who aim for it so maybe its too late if i propose that but who knows? ill try to propose it anyway...

(i really want to do database programming!!!! T__T)

This post has been edited by hazelrink: 26 November 2009 - 08:21 AM

[Momerath]

hazelrink, on 25 Nov, 2009 - 10:18 PM, said:

actually this is a thesis... sorry i didnt mention it...

Thesis as in Masters or PhD thesis? What time frame do you have to complete this thesis?

[hazelrink]

Momerath, on 26 Nov, 2009 - 09:43 AM, said:

hazelrink, on 25 Nov, 2009 - 10:18 PM, said:

actually this is a thesis... sorry i didnt mention it...

Thesis as in Masters or PhD thesis? What time frame do you have to complete this thesis?

actually we are just undergrad... we need to finish chapters 1, 2 and 3 of our thesis this sem and finish all th remaining chapters and defending it next year (4th year 1st sem).

[mostyfriedman]

I have a friend who already finished his thesis..it was about a DNA matching algorithm thing..

[hazelrink]

mostyfriedman, on 27 Nov, 2009 - 06:36 PM, said:

I have a friend who already finished his thesis..it was about a DNA matching algorithm thing..

how did they do that? (sorry i have no idea right now)

was the DNA's pictures and they made use of pattern matching algorithm?

I went to my prof and asked him if i could make a plagiarism detection tool (codes, paragraphs, etc) he said its good but i need journals and supporting literatures before he approves it... i just hope they approve it already next week!

[SwiftStriker00]

SpeedisaVirus, on 25 Nov, 2009 - 02:33 PM, said:

An actually effective bulk ID3 tagger and music organizer. It could reference online databases to propose possible tag data. Then if that is accepted you would then rename the files and organize as needed.

I've actually developed part of this, feel free to use it seeing how im sure i didnt use the best tactics. Also I posted the main algorithm on here as a tutorial

MP3 Renamer here
Search Algorithm here

Good luck on the project