k-means Cluster Shape Implications
| Autor: | Mieczyslaw A. Klopotek, Robert A. Kłopotek, Slawomir T. Wierzchon |
|---|---|
| Rok vydání: | 2020 |
| Předmět: | |
| Zdroj: | IFIP Advances in Information and Communication Technology ISBN: 9783030491604 AIAI (1) |
| DOI: | 10.1007/978-3-030-49161-1_10 |
| Popis: | We present a novel justification why k-means clusters should be (hyper)ball-shaped ones. We show that the clusters must be ball-shaped to attain motion-consistency. If clusters are ball-shaped, one can derive conditions under which two clusters attain the global optimum of k-means. We show further that if the gap is sufficient for perfect separation, then an incremental k-means is able to discover perfectly separated clusters. This is in conflict with the impression left by an earlier publication by Ackerman and Dasgupta. The proposed motion-transformations can be used to the new labeled data for clustering from existent ones. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|