Hierarchical Social Entropy 2.0

[Paul Johnson]

I've conferred with Tucker Balch and confirmed the calculation of HSE on
example data, so I'm ready now to claim "it works".  I've uploaded
HierSocialEntropy-2.0.tar.gz to the users-contrib/incoming directory on
the swarm ftp.  I've also got the goods here:

http://lark.cc.ukans.edu/~pauljohn/Swarm/MySwarmCode/HierSocialEntropy-2.0.tar.gz

HSE is a general purpose diversity measure. It relies on support classes
"AVLSet" (a Swarm wrapper around the AVL_Tree class in swarm's misc
directory) and "UDiamCluster" (short for U-diametric cluster algorithm)
which are also included.

To use this, all you have to do is include the classes in your source
code and Makefile, and then include HierSocialEntropy in your code, and
make a call on it like so:

ent = [HierSocialEntropy calculate: self  N: numOfObjects  DissMatr:
distMatrix MaxDist: max] ;

self = a Swarm memory zone, so you can do [self getZone] if you need to
numOfObjects = guess what?
distMatrix = a symmetric dissimilarity (distance) matrix of size
numOfObjects x numOfObjects
max = the largest possible distance between two objects.

In my previous HSE version, I had not allowed users to pass through the
maximum distance, but now I do, and, as I said, Tucker gets the same
numbers.

Tucker's been very kind in answering questions.

-- 
Paul E. Johnson                       email: address@hidden
Dept. of Political Science            http://lark.cc.ukans.edu/~pauljohn
University of Kansas                  Office: (785) 864-9086
Lawrence, Kansas 66045                FAX: (785) 864-5700

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.