Projects (Antoni Diller)

The following is a list of the areas in which I would be willing to supervise projects. If you are interested in a project in any of these areas, As I mentioned at the meeting about projects on Wednesday, the 30th of March 2011, I have a vary laid-back approach to supervising projects, favouring students who like to work on their own and who require a minimum amount of supervision.

Probabilistic games: I am willing to supervise projects which involve the use of computers in gambling (from the point of view of the punter). In other words, I am interested in programs which make money from games of chance. Some of the areas in which I've supervised projects in the past are: horse racing, Poker, Blackjack, the stock exchange, predicting Football results and betting on Snooker. I also supervised a project in which a student simulated a one-armed bandit or fruit machine.
Non-probabilistic games: I am willing to supervise projects involving any sort of game that can be played on a computer or that can be analysed with the help of a computer. Some projects from recent years that I have supervised include: Risk, Go, role-playing games, war games such as Axis and Allies, Hex, Ludo, Nine Men's Morris, adventure games, arcade-type games and Monopoly. I have also supervised projects in which people have written games for mobile phones.
Assertion evaluation: There is a huge amount of information available in electronic form on the Internet, but not all of it is correct. Similarly, not all the information that you read in books, in articles and in newspapers is correct. The idea of projects in this area is to attempt to automatically extract correct information from sources that may well contain errors or mistakes. Think about how you decide whether or not to believe what you hear or read (for example, "There are weapons of mass destruction in Iraq" as stated by Tony Blair before the invasion of Iraq): projects in this area involve automating this ability. Some work was done on this topic in the early days of AI by Abelson and also by Colby (of Parry fame), but only in recent years has their work been carried on. I supervised an excellent project in this area a few years ago which involved extracting information from a bulletin board about the possible transfers of players between clubs in the Premiership Football League and then deciding which of this information was correct.
Knowledge capture and extraction, text summarisation, etc.: There is a huge amount of information available in electronic form on the Internet. Search engines enable you to locate useful sites, but these may still contain a lot of information that is not relevant to your interests. I am interested in supervising projects which extract relevant information from electronic sources and maybe summarise this information or otherwise manipulate it in some way.
Functional programming language implementation: Bracket abstraction algorithms in combinatory logic have a use in the compilation of functional languages and I have several ideas about how to make such algorithms more efficient, but these new algorithms have only been implemented once so far. Projects in this area would involve implementing these algorithms.
The use of computers in psychotherapy: I am interested in projects which try to get computers involved in the process of counselling. This does not only apply to the programming of computers to play the role of a counsellor, but also the little-explored area of writing programs to help teach people how to be therapists.
LaTeX and TeX: I am willing to supervise LaTeX and TeX related projects. For example, projects which involve writing a set of TeX macros to rival Lamport's LaTeX or writing a package for the typesetting of film scripts.
Writing stories by computer: I've supervised projects about getting computers to write stories and am interested in supervising further projects in this area.
Off-the-wall projects: If you have any whacky or way-out ideas then come and see me. I may well be interested in supervising your off-the-wall project.