Searching for a masters thesis topic

Question

Hello!

I am a student at an university and I have to write my master thesis (information technology / computer science). I have experience with programming (Objctive-C, PHP, web development, Java, C/C++, JSP, J2ME) and SQL databases like MySQL, PostgreSQL, Apache Derby.

I am currently interested in two topics: visualization and/or security.

Security: I have knowledge in crytography, network systems, firewalls, I did some basic implementations of RSA and AES. But I have no idea what is interesting in the future, I do not have the needed overview of "what's next".

Visualization: I found the project Sixth Sense of the MIT Media Lab (http://www.youtube.com/watch?v=eUy8j4TMH2M) which is a wearable device with a projector and a camera which recognize finger gestures and so you can "control" the user interface. I found this very interesting because in my opinion mobile devices will be more powerful (better processors, longer battery life, ...) in the future and it would be nice to have the screen instead of a display on every surface you want.

Because I only have 6 months to do my thesis (2-3 months for developing something) and so I only can build up a prototype and I am searching for ideas. Perhaps it is possible to do an iPhone app?!

Perhaps someone could give me some suggestions.

Thanks a lot in advance & Best Regards.

Answer 1

Having just finished my MSc, I'll offer what advice I can.

• Get a good supervisor (by good I mean someone you can talk to easily and is interested in similar areas)
• Start early - just because there's an 'official' start date doesn't mean you can't get a head start, especially with planning.
• KEEP A NOTEBOOK. 6 months was the longest academic project I had ever done. I did the majority of the writing in the last month where the notebook was invaluable.
• If you're interested in it, you'll work on it (probably night and day!), so pick a topic you're enthusiastic about.
• Make sure you're performing some kind of novel research/ answering some kind of question. At least at my Uni (in the UK), this was a criteria for a good masters project. Simply making something cool might not be enough.

Answer 2

You could always comboine your two interests and do something with security visualization.

Answer 3

You should take merge the ideas of Security and Visualization. For instance currently Capthcas are quite weak. Vulnerabilities have been reported in catpcha implementations, but not many people are looking: http://web.nvd.nist.gov/view/vuln/search-results?cid=2

Being able to develop a successful attack against a Capthca would be excellent material for a thesis. I recommend looking at open source capthca's such as Jcaptcha.

A known attack method is to train an OCR using known captcha images to increase the odds of reading unknown images. http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract