Substitutability by counting linkages (Google seems to be of no help)

S. Dale Morrey sdalemorrey at gmail.com
Sun Jan 22 19:21:00 MST 2012


Hi Everyone,

I'm hoping someone can help me remember the name for an idea I've
stumbled across.
I've heard of this idea before but now I can't think of the name and I
would like to include some sort of reference in a paper I'm writing.

The crux of the idea is that you can predict the likely hood of a
substitution occurring by counting the common linkages between two or
more items.

For instance, if I'm in a grocery store and I know I need to buy meat
and I have three choices which are, beef, chicken and pork.
All of them would link up together at meat, however beef and pork also
link up under mammal meat, therefore beef and pork would be more
adequate substitutes for one another than chicken would be for either
of them.

On an inheritance graph you might see that beef, pork and chicken are
all inheriting from meat.  However beef and pork would also inherit
many traits from mammal, where as chicken would inherit from bird.
Since the inheritance occurs further down the graph, for beef and
pork, they are more closely tied and one could predict that a person
is more likely to pick beef or pork as substitutes for one another
than they would for chicken.

It's driving me nuts, I'm sure this has a name and I think I remember
reading about it years ago, but for now I don't seem to be able to
find anything on it.

Does anyone have any ideas?


More information about the PLUG mailing list