Hello
I've got to choose a project to do for my 3rd year at Uni. I'm currently looking at these two:
Patience Players Advisor
Failure to get a game of patience to come out might be due to lack of skill on the player's part. But it might simply be due to a bad lie of the cards. It would be nice to have a Patience Players Advisor which, before play, analysed the hand dealt and was able to indicate whether or not the hand was playable to a successful conclusion. The advisor might also suggest the number of moves needed. You might start with a single game of patience say Canfield and then generalise your solution to handle a wider range of patience games.
and
Solving the Times Crossword puzzle using search techniques
The aim of this project is to develop a program for solving the times crossword puzzle which constructs a search space from the dictionary and the grid, and then uses some form of dictionary driven search for solutions to intersecting partially solved sections of the grid. One approach might be to drive the search by some notion of match between a clue and a candidate answer. For example, you might use a genetic algorithm as the search technique, and your notion of match might be whether the crossword clue contained one or more synonyms of the answer. One question to study is how the size of the search space reduces as an increasing portion of the grid is completed. To obtain good results it will probably be necessary to help the mechanical search by giving it information about the type of clue (anagram, run-on etc), and access to a thesaurus. You could link your system to one of the many online English dictionaries or thesauri (e.g. http://www.dictionary.com; http://www.lightlink.com/bobp/wedt).
Does anyone have any opinions on which would be good to do, if they'd be impossibly hard...etc.
Any comments would be fantastic! Thanks, Paul
|
