next up previous contents
Next: Algorithm Up: Exploiting DASSS Data in Previous: Maximizing for : DASSS   Contents

Maximizing $p(u,v,r\vert D)$ for $D$: DASSS Data

Armed with this set of $p(D\vert u,v,r)$ values and the independently considered $P(u,v,r)$, we may finally evaluate $P(u,v,r\vert D)$ using Bayes' rule as suggested above,

$\displaystyle P(u,v,r\vert D) = \ensuremath{\frac{p(D)\vert u,v,r)p(u,v,r)}{p(D)}}.$     (27)

Then, determining $(u,v,r)$ to satisfy expression 14 is simply a matter of searching over the $N$ choose 2 values for the optimal one.



Aaron S. Master 2003-11-01