As I previously mentioned, I am using log(N) as the measure of ambiguity for a concept, where N is the number of meanings of the concept. The ambiguity of a sentence, or a hierarchy, is the sum of the concept ambiguities.
You can use mKE to calculate these ambiguities for you. When mKE reads sentences, it collects all the genus-species-unit information and stores it in a single table: h[concept] := GSU(g,s,u); h is the hierarchy name GSU is the record consisting of g is set of names of all genera of concept s is set of names of all species of concept u is set of names of all units of concept The concept ambiguity is the log of the size of the genus set. a[concept] := log(*g);
The typical ontology hierarchy is a lattice which includes all N definitions of a concept. For instance, in John's recent email, NLP has two meanings, and a[NLP] := 1 bit. Hopefully, when we integrate all the available context for a sentence, the context lattice reduces to a tree with a[h] := 0 bits and a[sentence] := 0 bits.
In mKE, you use the ambiguity command to calculate the ambiguity of a concept, sentence or hierarchy: a[concept] := do ambiguity od concept in h; a[sentence] := do ambiguity od {sentence;} in h; a[h] := do ambiguity od h;
Dick McCullough Context Knowledge Systems What is your view?
|