Data cluster analysis: Abrupt transition
A cluster analysis optimization technique for partitioning data is applied to a class of distributions which are functions of a single variable. It is shown that under certain realizable conditions an infinitesimal change in the data distribution causes a large, abrupt change in the location of the optimal boundary separating two data clusters. This abrupt (phase) transition is illustrated with an example.
Year of publication: |
1978
|
---|---|
Authors: | Falk, H. ; Falk, C.T. |
Published in: |
Physica A: Statistical Mechanics and its Applications. - Elsevier, ISSN 0378-4371. - Vol. 90.1978, 2, p. 325-332
|
Publisher: |
Elsevier |
Saved in:
Saved in favorites
Similar items by person
-
Random mapping based on a genetic model
Falk, H., (1979)
-
Remarks on stochastic non-linear birth and death models
Falk, H., (1980)
-
Glauber's stochastic spin model with broken bond disorder
Falk, H., (1983)
- More ...