geva.hcluster {geva} | R Documentation |
Performs a hierarchical cluster analysis from summarized data.
geva.hcluster( sv, resolution = 0.3, hc.method = options.hc.method, hc.metric = options.hc.metric, cl.score.method = options.cl.score.method, ..., include.raw.results = FALSE ) options.hc.metric # c("euclidean", "maximum", "manhattan", "canberra", # "binary", "minkowski") options.hc.method # c("centroid", "median", "ward", "single")
sv |
a |
resolution |
|
hc.method |
|
hc.metric |
|
cl.score.method |
|
... |
additional arguments:
|
include.raw.results |
|
This function performs a hierarchical cluster analysis with the aid of implemented methods from the fastcluster::fastcluster
package, particularly the fastcluster::hclust.vector()
function. The available methods for the hc.method
and hc.metric
are described in the function's documentation page (see fastcluster::hclust.vector()
).
The resolution
value is an accessible way to define the cluster separation threshold used in hierarchical clustering. The algorithm produces a dendrogram-like hierarchy in which each level/node is separated by a distance (sometimes called "height") to the next level/node, and the resolution
translates a value between 0
and 1
to a propotional value within the total hierarchy height. This allows defining the rate of clusters from 0
to 1
, which results in the least number of possible clusters (usually two) for 0
, and the highest number (approximately one cluster per point) for 1
.
If include.raw.results
is TRUE
, some aditional data will be attached to the info
slot of the returned GEVACluster
objects, including the kNN tree generated during the intermediate steps.
A GEVACluster
object
In hierarchical clustering, all points are clustered. Therefore, setting resolution
to 1
will result into one cluster per point, where the cluster analysis may become pointless (no pun intended).
Other geva.cluster:
geva.cluster()
,
geva.dcluster()
,
geva.quantiles()
## Hierarchical clustering from a randomly generated input # Preparing the data ginput <- geva.ideal.example() # Generates a random input example gsummary <- geva.summarize(ginput) # Summarizes with the default parameters # Hierarchical clustering gclust <- geva.hcluster(gsummary) plot(gclust) # Hierarchical clustering with slightly more resolution gclust <- geva.hcluster(gsummary, resolution=0.35) plot(gclust)