next up previous
Next: Network Evaluation Up: Theory: Evidence and Association Previous: Identity Inhibition

Evidence Integration

There are eight evidence types, as discussed in the previous eight sections, and a single integrated plausibility value needs to be computed from them. All values are assumed to be on the same scale so this simplifies the considerations.

Some constraints the computation should meet are:

Based on these constraints, the following integration computation has been designed:

Let:  
  $evd_{prop}, evd_{desc}, evd_{subcl}, evd_{supcl}, evd_{subc},
evd_{supc}, evd_{ass}, evd_{inh}$
   
be the eight evidence values, with weightings:
  $w_{prop}, w_{desc}, w_{subc}$
Then:  
  $v_{1} = harmmean(\{(evd_{prop},w_{prop}),(evd_{desc},w_{desc}),(evd_{subc},
w_{subc})\})$
  if $evd_{supc} > 0$
  then $v_{2} = v_{1} + c_{supc}*evd_{supc}$ $(c_{supc} = 0.1)$
  else $v_{2} = v_{1}$
  if $evd_{ass} > 0$
  then $v_{3} = v_{2} + c_{ass}*evd_{ass}$ $(c_{ass} = 0.1)$
  else $v_{3} = v_{2}$
  if $evd_{supcl} > 0$
  then $v_{4} = v_{3} + c_{supcl}*evd_{supcl}$ $(c_{supcl} = 0.1)$
  else $v_{4} = v_{3}$
  if $evd_{inh} > 0$
  then $v_{5} = v_{4} + c_{inh}*evd_{inh}$ $(c_{inh} = -0.25)$
  else $v_{5} = v_{4}$
   
Finally, the integrated plausibility is:
  $min(max(v_{5},evd_{subcl},-1.0),1.0)$
   

The $-1.0$ and $+1.0$ terms in the final function ensure the result is in the correct range. The weighting constants (0.1 and $-0.25$) used above were chosen to influence but not dominate the evidence computation and were found empirically. Small changes (e.g. by 10%) do not affect the results.

The invocation network fragment executing this function is similar to those previously shown except for the use of a "gated-weight" function unit that implements the evidence increment function for supercomponent, association, superclass and inhibition evidences.


next up previous
Next: Network Evaluation Up: Theory: Evidence and Association Previous: Identity Inhibition
Bob Fisher 2004-02-26