Statements (27)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:algorithm
|
| gptkbp:advantage |
scalable to large datasets
better cluster quality than standard K-means |
| gptkbp:application |
document clustering
text clustering |
| gptkbp:complexity |
O(nkt)
|
| gptkbp:differenceFromKMeans |
not all clusters split simultaneously
splits one cluster at a time uses divisive approach |
| gptkbp:firstPublished |
2000
Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining |
| gptkbp:input |
gptkb:dataset
number of clusters |
| gptkbp:output |
cluster centroids
cluster assignments |
| gptkbp:proposedBy |
Steinbach, Karypis, and Kumar
|
| gptkbp:step |
applies K-means with k=2
repeatedly splits clusters into two |
| gptkbp:supportsAlgorithm |
hierarchical clustering
divisive clustering |
| gptkbp:usedFor |
cluster analysis
|
| gptkbp:usedIn |
gptkb:machine_learning
data mining |
| gptkbp:variant |
gptkb:K-means_clustering
|
| gptkbp:bfsParent |
gptkb:K-means_clustering
|
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
Bisecting K-means
|