I need to look into this in some detail before I can determine the usefulness of your proposal. 1) I suspect that further investigation might lead to better characterization of F. 2) I have to figure out how to interpret L and W as you have suggested since I was looking at scalar values originally.
Anyways, thanks for this suggestion Raul. On Fri, Feb 26, 2010 at 12:27 PM, Raul Miller <[email protected]> wrote: > On Fri, Feb 26, 2010 at 2:42 PM, Yuvaraj Athur Raghuvir > <[email protected]> wrote: > > 0: l1 F w1 (Outdoors - Sunny) > > 1: l1 F w2 (Outdoors - Rainy) > > 2: l2 F w1 (Porch - Sunny) > > 3: l2 F w2 (Porch - Rainy) > > 4: l3 F w1 (Indoors - Sunny) > > 5: l3 F w2 (Indoors - Rainy) > > > > and > > > > (l1 F w1) > (l2 F w1) > (l3 F w2) > (l3 F w1) > (l2 F w2) > (l1 F w2) > > With this characterization, we could define > l1=: Outdoors=: 1 0 0 > l2=: Porch=: 0 1 0 > l3=: Indoors=: 0 0 1 > > w1=: Sunny=: 1 0 > w2=: Rainy=: 0 1 > > M=: 3 2$0 > g=: <@,&I. > NB. M=: M (l1 g w2)}~ 0 NB. unnecessary > M=: M (l2 g w2)}~ 1 > M=: M (l3 g w1)}~ 2 > M=: M (l3 g w2)}~ 3 > M=: M (l2 g w1)}~ 4 > M=: M (l1 g w1)}~ 5 > F=: +/ .* M&(+/ .*) NB. or F=: {&M@<@,&I. > > L=: l1,l2,:l3 > W=: w1,:w2 > > With these definitions: > L F"1/ W > 5 0 > 4 1 > 2 3 > > Which should not be very surprising because > M > 5 0 > 4 1 > 2 3 > > and L and W are identity matrices. > > If F (or L or W) need to have other qualities, > this might not be a good approach. So I have > to ask: is this useful to you? > > Thanks, > > -- > Raul > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
