|
gptkbp:instanceOf
|
gptkb:algorithm
|
|
gptkbp:application
|
image segmentation
document clustering
market segmentation
vector quantization
|
|
gptkbp:category
|
unsupervised learning
cluster analysis
|
|
gptkbp:complexity
|
O(nkt)
|
|
gptkbp:convergesWhen
|
cluster assignments do not change
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
K-means
|
|
gptkbp:implementedIn
|
gptkb:MATLAB
gptkb:scikit-learn
R
|
|
gptkbp:input
|
set of data points
|
|
gptkbp:introduced
|
gptkb:Stuart_Lloyd
|
|
gptkbp:introducedIn
|
1957
|
|
gptkbp:limitation
|
not suitable for categorical data
assumes spherical clusters
sensitive to initial centroids
|
|
gptkbp:measures
|
Euclidean distance
|
|
gptkbp:objective
|
minimize within-cluster variance
|
|
gptkbp:output
|
set of clusters
|
|
gptkbp:relatedTo
|
gptkb:Gaussian_Mixture_Model
gptkb:DBSCAN
hierarchical clustering
|
|
gptkbp:requires
|
number of clusters (k)
|
|
gptkbp:step
|
assign points to nearest cluster center
update cluster centers
|
|
gptkbp:supportsAlgorithm
|
unsupervised learning
|
|
gptkbp:usedIn
|
gptkb:machine_learning
data mining
|
|
gptkbp:variant
|
gptkb:K-means++
gptkb:K-medoids
gptkb:Mini-batch_K-means
|
|
gptkbp:bfsParent
|
gptkb:DBSCAN
|
|
gptkbp:bfsLayer
|
6
|