Ah, now this is more like it. The sun is shining, I got up at a… - John C. Kirk
Dec. 9th, 2002
Ah, now this is more like it. The sun is shining, I got up at a reasonable time (someone phoned at about 10:30), and there some interesting MSc projects up on the Kings website. This is shaping up to be a pretty good day so far :)
Now, some of the projects look better than others, it must be said. One that seems overly ambitious is "Voice recognition over the Internet. As each person's voice is unique, it can be used for security applications, like iris recognition. The project would be to write software in Java to enable on-line verification eg. at airports, in on-line banking. This project would be in collaboration with Securicor Ltd." You have 3 months - good luck! I'll give that one a miss, I think...
However, the list on this page look good, particularly the bottom one. This is actually the standard example I use when I talk to people about Natural Language Understanding (the area I'd like to do a PhD in), as a case where you need some understanding rather than just translation. Suppose you have a chunk of text like "Peter picked up the ball. He gave it to Bob." The computer needs to understand that "he" in the second sentence means "Peter", and "it" means "the ball". You could then ask the computer "who has the ball?" and it should say "Bob". Basically, the type of thing you do in English comprehension at primary school. I heard about this kind of thing (as a problem) when I was doing my undergrad degree in Durham, but I never went into it in any more depth than that. So, if I can get a project like that, then it would be interesting (which would stop me moaning on here all the time about being bored!), and would be a good stepping stone for a PhD, particularly since they say that it will extend an existing system at Kings (sounds like their version of Durham's LOLITA project). A minor problem is that I've never used PROLOG before, but I'm sure it can't be that hard... Hmm, actually, just had a brief moment of panic, when I saw that the list I mentioned is for 2001/2002 (the department website isn't always as up to date as it could be...) However, there's an identical list here (although with less descriptions, which is why I'm leaving the link above), so it looks like they're still available. And in fact, if that means that nobody did them last year, then I should have a good chance of getting one this year. The main obstacle is likely to be that I'm doing the MSc in Advanced Computing, rather than the MSc in Natural Language Processing, so that probably means that people on the other course would get "first dibs". Still, can't hurt to try. Actually, the guy running those projects (Shalom Lappin) is one of the people who replied to me when I was trying to go straight into a PhD (back in April), so that bodes well, i.e. he may be sympathetic. I'll drop him an email, to enquire further. Stay tuned for updates :)
I was also amused by the descriptions of grade boundaries here: "Perfect, excellent, very good, good, quite good, satisfactory, relatively poor, poor, very poor, abysmal, worthless". Reminds me of Dilbert strips - "for your appraisal, I rate you as dung beetle".