You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The cluster analyses allow the user to assess regional clustering of similar
values, but using a set number of indices that have been enabled for
clustering.
A region grower analysis would allow the user to use any of the indices,
maximising or minimising the selected index as appropriate.
This should be limited to those indices that "lump" groups in the
neighbourhoods together so the matrix is guaranteed to be symmetrical.
Otherwise indices like ENDC_CWE will produce a matrix entry for [group1,
group2] that is not the same as for [group2, group1].
This can be developed as an extension to the clustering analyses, as all that
is needed is to allow a greater set of indices be used for the clustering, and
maximise the values by default. The more tedious part is to add metadata for
all the relevant indices.
Original issue reported on code.google.com by shawnlaffan on 16 Jan 2011 at 12:18
The text was updated successfully, but these errors were encountered:
Original issue reported on code.google.com by
shawnlaffan
on 16 Jan 2011 at 12:18The text was updated successfully, but these errors were encountered: