VMs: Pair distributions

["J HALEY" <j.haley@xxxxxxxxxxxxxx>]

Hi

I have just posted two files to the george boole site. Here are
the links.

http://www.georgeboolefanclub.com/latinpairdistribution.txt
http://www.georgeboolefanclub.com/vmspairdistribution.txt

Each file was created using an algorithm that scanned the
text recording each pair of elements it found in a table. The
elements appear in the list in appearance order. That is,
taking the first line in the VMS as an example, 
(f a ch y s.y k al) the first entries in the table would become:

.f a 1
a ch 1
ch y 8
y s. 1
.y k 55

Here there are 55 occurances of the pair y k, the dot indicating
that these are only recorded at word starts. I did this for both
Latin and the VMS and distinct popularity bands appear relatively
early with less popular pairings trailing at the end. The two files
show enough similarities to indicate that the probability of meaning
in the VMS is quite high. The more distinct bands of high frequency
pairs also show a tendency to group together in the Latin. I would
be interested in any views on this

Jeff


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list