Belief Operations

David Fisher

Various belief operations tell the belief nodes in the inference network how to score (or if to even score) various terms. The belief operations are:

Belief Operations:

name Example
combine #combine(dog train) 0.5 log( b(dog) ) + 0.5 log( b(train) )
weight, weighted and #weight(1.0 dog 0.5 train) 0.67 log( b(dog) ) + 0.33 log( b(train) )
weighted sum #wsum(1.0 dog 0.5 dog.(title)) log( 0.67 b(dog) + 0.33 b(dog.(title)) )
not #not(dog) log( 1 - b(dog) )
max #max(dog train) returns maximum of b(dog) and b(train)
or #or(dog cat) log(1 - (1 - b(dog)) * (1 - b(cat)))


Wiki: Home
Wiki: The Indri Query Language


  • ChanBen


    Can you tell me please, what does mean the "b" in each equation ??
    And can we consider the "#combine" as an "AND" operator ??

    Last edit: ChanBen 2015-05-01
  • David Fisher
    David Fisher

    b() is the belief operator. Typically an estimated probability. #combine is an OR operator