>From: <[email protected]>
>To: <[email protected]>
>Sent: Friday, November 07, 2008 4:39 PM
>Subject: [quackle] Digest Number 525


> 3a. Re: Estimating vocabulary advantage...
>    Posted by: "Kevin Leeds" [EMAIL PROTECTED] vekkus4
>    Date: Thu Nov 6, 2008 6:16 pm ((PST))
>

> Human errors might be classified somehow, and Quackle could be programmed
> to try to mimic them.  Possible categories:
>  Don't know a word at all
>  Know a word, but only x % of the time (x is between 0 and 100).
>   (This would take a lot of research to develop)
>  Not very good at using hot spots yet
>  Too afraid of setting up bingo hooks
>  Too afraid of setting up hot spots on triples (both red & blue)
>  Too fond of using 3 letter words like COW
>  Noplays the opponent when more than 3 bingos are made
>  Fails to notice the number of tiles in the bag
>
> I think the list is both incomplete, yet too long now.

I don't think it is too long.

To simulate a user who doesn't know the whole lexicon, make Quackle play only 
words from a smaller lexicon. There are such word 
lists on the web, e.g. here:
http://wordlist.sourceforge.net/
http://aspell.net/
Such word lists could of course be enhanced by the addition of words which a 
novice would most likely learn first, e.g. the 2s, 3s 
and some of the likeliest 7s.

For every thing that Quackle puts a value on, could Quackle be made to mimic a 
human player who has the weakness of either 
overestimating or underestimating it? For example:

Underestimates the importance of good leave
Overestimates the importance of good leave (bonus-chaser who would rather play 
off one tile for a low score than make a play which 
gets a good score and still has a fairly good leave)

or, more generally, "uses an inferior evaluation of leave". We could, for 
example, have a user who:

Underestimates a blank, an S or the Z
Overestimates the Q
Assesses leave  merely on vowel/consonant balance 


Reply via email to